Finally have a pick and place on the way Quad IVC

[rx8pilot]

I think this will be the beginning of a journey, so I thought I would start a thread for it. I just purchased a Quad IVc with 55 feeders to help boost my small operation. This is to help me move out of hand placing which I have been doing for too long now. Most of my PCB's are dense and double sided. Some with 300+ components and fine pitch. The boards seem to change a lot, so I have to do them in small batches which is slow and expensive for contract manufacturing.

The Quad IVC is a 90's era workhorse that runs on MS-DOS software. Fortunately, I am well versed in the DOS world so that by itself is not a problem. The old computers and custom electronics have me crossing my fingers a little. Obviously the goal is to save time, not spend it all fiddling with a machine. The Quads are very well supported by PPM (at a cost of course). There are many upgrades and spares available if/when the need arises.

I have poured over every imaginable solution to pick_n_place for low budget operations. DIY, Madell, SMTmax, TM22A/TM240A, etc, etc. Many hours spent weighing the pro's and cons. My thinking up until today was to get buy the LitePlacer DIY kit https://www.eevblog.com/forum/manufacture/pick-and-place-machine-kit-with-vision/ to get my prototypes and pilot runs done. After that, the goal was to buy a Quad machine from PPM that is totally updated and warranted for one year. The Quad machines are designed like tanks and can run continuously and have room for large numbers of feeders. A quad will certainly carry all of my pick and place needs for quite some time (if it works). There are also enough feeders to have the machine setup for ALL of PCB's at the same time.

At the price that I paid, I can repair and replace a lot of expensive parts and still come out ahead. The hope is that it works reasonably well right away. We shall see.
If it can get through the first 100 sets of PCBs, the cost will have been covered relative to the contract manufacture costs. Hopefully I can post some pics and write a little about the experience of finding and buying a PnP machine that is intended to make money. For this to work, it has to be a great balance of low initial cost and reasonable long-term fiddling.

[Corporate666]

Got your PM - will send my contact info over your way.

I'm very curious to keep up with your progress thread.  I'm guessing you bought the one on eBay with the PPM software upgrade?  If so, you got an amazing deal... I didn't even see that on eBay and if I had, I'd have bought it

I am not familiar with the PPM software - I just saw a demo of it, but haven't used it.  But it uses the same base machine, it's just an upgraded "main board" and software.

Where in the country are you?  I'm not too far from PPM - but honestly, they charge (IMO) absolutely outrageous prices so I use them as little as possible, so if you're nowhere near that's no problem.

Assuming you got the eBay one, it looks like the table has been modified for a specific use.   That's one of the big pro's of these machines...they are very configurable.  It looks like the conveyor rails have been moved 6-10 inches off to the right (or maybe even shortened?).  Do you know how boards load - do they feed both into and out of the right side?  That would be something you could set up in the PPM software, I guess.

Be sure you level the machine well (ideally, use a machinists level, not one from the hardware store). The more unlevel the machine is, the more it's working harder in one direction than another. 

Anyway, great buy - interested in how things go

[rx8pilot]

That is the one.....eBay super special and it's only a short drive for me to pick up.

Can't tell what the table setup is like. It looks like a shuttle system of some sort. I was expecting to machine my own fixtures to hold my boards since none of them are square. Fortunately, my previous career was a mechanical engineering and machining so I can make anything that a 5 axis CNC can make quickly. I also have a serious machine level - .0001"/foot which should be workable.

Pretty happy with this deal, there are a lot of feeders including some of the bigger ones for 24mm, 36mm, and 44mm. I don't see any of the reel arms so I may need to pick those up from PPM. Still, a rough estimate of the feeders is north of $25k I think.

Do you know of any online manuals? Not the PPM ones, but the originals? Glad you got the message, look forward to seeing what this can do for my business.

[Corporate666]

That is the one.....eBay super special and it's only a short drive for me to pick up.

Can't tell what the table setup is like. It looks like a shuttle system of some sort. I was expecting to machine my own fixtures to hold my boards since none of them are square. Fortunately, my previous career was a mechanical engineering and machining so I can make anything that a 5 axis CNC can make quickly. I also have a serious machine level - .0001"/foot which should be workable.

Pretty happy with this deal, there are a lot of feeders including some of the bigger ones for 24mm, 36mm, and 44mm. I don't see any of the reel arms so I may need to pick those up from PPM. Still, a rough estimate of the feeders is north of $25k I think.

Do you know of any online manuals? Not the PPM ones, but the originals? Glad you got the message, look forward to seeing what this can do for my business.

Are you getting your PCB's made individually, as opposed to panelized?  What is the PCB thickness you are using?

I always get my PCB's panelized with 0.3" rails on 2 sides, and two fiducial marks on each side of the panel.  What are your boards like in terms of parts?  Is it mostly larger stuff or are you doing any fine pitch (QFN, etc)?  If your boards aren't setup for pick-and-placing, you might find you will need to either make some compromises or if you have some high volume boards, it may make more sense to respin some of those boards. 

It seems like we have really similar setups - we do CNC machining as well as electronics here.. for my previous PnP, I machined carriers out of 0.125" 6061 aluminum.  I would cut a 0.062" pocket in the sheet to hold the PCB, and I would machine a 0.3" wide 0.062" slot on either side of the carrier.  This made it easy to hold in my conveyors.  If you have holes in the PCB you could use dowel pins to hold them instead of routing out shapes for them - but you will do much better if you have features the PnP can use for registration.  I believe you can use pads or holes for registration - the manuals tend to strongly advise against this, so it may not work well or isn't reliable - I dunno. 

Looking at the pics of your machine, if you don't want (or choose not to) relocate the conveyor, you could remove the waffle tray and machine two plates.  A base plate out of 0.25" mic plate with some reamed holes for locating pins, then a top fixture plate with reamed holes that will be removable and sit on top.  Then just machine that top plate with pockets for your PCB's.  Hell, you could even machine pockets next to the PCB pockets for chips necessary for the PCB if you like   Nice thing about the 4C top is that it's just an array of tapped holes all across the table, so anything can be moved.

Going forward with new PCB's - traditionally there are 2 fiducial marks on each side of the PCB panel that the PnP will read and use.  By measuring where those marks are vs. where it thinks they should be, it can correct for board position errors like X/Y offset, rotational skew and even elongation in X/Y axes.  This gets important the smaller your parts are.  You can also put local fiducials on tricky parts like QFN's or BGA's (you've probably seen fiducials on either corner of a fine chip).

http://www.mvdmoosdijk.nl/Electronics/Art/Fiducials/fiducial_panel.gif



I have a full set of manuals, I can copy them for you next time I go to Kinkos or something.  They would only be partially applicable though since we have different software.  The hardware part of the manuals would still apply as your machine uses the same motor drivers as mine, but your main software is different. 


Do you still have your CNC machines in the shop?  And still have a 5-axis machine?  Maybe you can help me out with something?   I have a part I am working on that is a dodecahedron (12 sided polygon, each side is a pentagon).  I need to machine it out of 6061, and each face has a few tapped holes and a deep drilled hole.  We're in the process of changing some machinery here so I no longer have a 5-axis machine.  I was going to try to maching it on my 3-axis VMC with a tilting rotary indexer, but if you have a 5-axis machine.... maybe you could help?

[rx8pilot]

Are you getting your PCB's made individually, as opposed to panelized?  What is the PCB thickness you are using?

At the moment all PCB's are individually routed because they mainly manually placed. Over the past few months, the designs have really stabilized so I feel more comfortable with bigger runs. This PnP purchase was definitely an impulse buy because it seemed like a good deal. Now that I have a production level machine, penalization will be important. I still need to do prototypes with odd shapes, so something needs to be created for those.

Going forward with new PCB's - traditionally there are 2 fiducial marks on each side of the PCB panel that the PnP will read and use.  By measuring where those marks are vs. where it thinks they should be, it can correct for board position errors like X/Y offset, rotational skew and even elongation in X/Y axes.  This gets important the smaller your parts are.  You can also put local fiducials on tricky parts like QFN's or BGA's (you've probably seen fiducials on either corner of a fine chip).

Yes there are QFN's at .5mm and hopefully BGA in the future. Do you think I should use local fiducials for those parts?

I have a full set of manuals, I can copy them for you next time I go to Kinkos or something.  They would only be partially applicable though since we have different software.  The hardware part of the manuals would still apply as your machine uses the same motor drivers as mine, but your main software is different. 

Thank you, I should be patient. It looks like there are some manuals included that may be enough to get me started.

Do you still have your CNC machines in the shop?  And still have a 5-axis machine?  Maybe you can help me out with something?  

For just a little longer. The CNC's are going away soon. I should still be able to help with a one-off. The 5-axis makes weird parts easy - totally changed the way I design once the 5 axis capability arrived.

[Corporate666]

Yes, you should absolutely try to do local fiducials for BGA, certainly... for QFN, you are probably OK without them, but it won't hurt.

I would guesstimate that my place accuracy is reliably about < 0.010", and that is just with panel fiducials and nothing on each individual PCB or on any chips.  That should be fine for even 0.5mm fine pitch (0.5mm being 0.020"), but using the upward camera and local fiducials will get you down to 0.005" or likely less.  Make sure you have the right nozzles as well.  If you have an NC lathe, you can make them pretty easily - some of them have rubber attachments around the chip which help seal the vacuum in and prevent heavier parts from dropping.

On the feeders - I have found them to be generally reliable.  There are occasionally one or two that will error-out but the machine will retry multiple times and then beep at you if it can't pick the part, so it's not like you will get a board with bad placements or missing parts - it just means you will need to go to the machine and fix whatever is wrong.  I've had very mixed luck in fixing feeders.  Sometimes the problems are obvious - especially fluff/debris from paper tapes accumulating in the gear mechanisms.  In such cases, compressed air and oil helps a lot.  But in other cases, the feeder works fine 90% of the time and 10% of the time doesn't.  Getting them just right is sort of like getting a mountain bike adjusted just right... it's a bit of an art to get it perfect.  PPM (and I am sure others) offer a refurb service - I'll probably send my few problem feeders in for repair/replacement at some point.

If you are ever buying more feeders, get 12V feeders.  The 12V indexes much faster... 5V much slower.  The machine can actually arrive back at the feeder to pick the next part, and a 5V feeder might not have indexed fully yet, slowing things up (only relevant when you are picking multiple parts in succession from the same feeder).

If you do end up needing manuals, let me know and I will copy 'em.

On the 5-axis part... I had an idea of how to make the part from a few separate pieces, so I'll give that a shot first - and if not, I might come begging for help

[marshallh]

1.0mm BGA is very forgiving of part placement. The chip floats on the surface tension of molten solder, they "pop" into place much more than any other package. Compare to 0.4,0.5mm QFP/QFN which is killer.
My fab said they don't need fiducials for BGAs (they just use the ball lands themselves), and fiducials for other parts are nice but not required (of course, completely different equipment..)

[rx8pilot]

1.0mm BGA is very forgiving of part placement. The chip floats on the surface tension of molten solder, they "pop" into place much more than any other package. Compare to 0.4,0.5mm QFP/QFN which is killer.
My fab said they don't need fiducials for BGAs (they just use the ball lands themselves), and fiducials for other parts are nice but not required (of course, completely different equipment..)

Yes, the .4mm QFN's I am using now are not forgiving at all. I am REALLY looking forward to being able to use BGA's and small passives to improve density and performance of my designs. I think I will need special feeders if I go to 0201 so I may stick with 0402 passives.

Picking up the system on Thursday AM, fingers crossed that it is all there and makes the journey safely.

[Kjelt]

Great topic I think many enthousiasts here, as I am, are excited to see how this will develop.
I always wonder what kind of products you make/sell that you can upgrade to your own productionline.
And I am also interested to see the rest of your line like the stencil paste machine, flying probe tester and reflow oven ?

[rx8pilot]

I hope so, the overall cost of the Quad IVC with feeders was less than a TM240A which is FAR less capable. For an enthusiast its a chunk of change, but not out of the ball park. For me, this is a transition from high-end hobby to a true business. I have been making money from electronic design for a while, but it has all been very basic and hand assembled designs. My main income has been mechanical design and manufacturing, now it will include electronics as well. My entire PnP 'line' is well under $10k and in theory can make more boards in a day than I could hand assemble in a month with better consistency.

I purchased a second hand JEM-310 convection batch oven about 2 years ago and it has been great for what I do. Far better than a DIY toaster oven, but it was about $3k or so. It can manage lead free solder and can do one panelized PCB at a time. The stencil printer started out as a Manncorp 2100, but I added my own machined PCB holders to make it work for fine pitch. The way is comes from Manncorp is useless for almost anything at all. This PnP 'line' is about as basic as it can get and about as cheap as it can get and still be considered professional. I only need to make a few hundred PCB's per month right now so it will be a leisurely pace for this setup. I really wanted the control and oversight of the builds to avoid mistakes and correct any issues on the fly. This will also give me a crash course education on DFM - Design for Manufacture.

[rx8pilot]

Ok, here we go. Today I went to pick up the machine. I learned from PPM (Precision Placement Machines) that it is now called a Quad 4000C since they upgraded it to windows from its original DOS configuration. I have not done a proper inventory of the feeders, but there are 55 total which is a great start for a PnP. 8mm, 12mm, 16mm, 24mm, waffle trays, and a 5 lane tube feeder.

It needs a serious cleaning for sure, probably about a full day or more of just cleaning it up which will give me a chance to learn what is what. I did boot it up but there was some issue with the software license and it would not really do anything. I spoke with the people at PPM (goppm.com) and they were very generous, spending about an hour on the phone helping me inspect the machine reasonably well. It will probably take a few days to actually get it running with all the cleaning and installation tasks. It also has a PCB conveyor but I don't think I will use it. I need to design and build a general purpose PCB holder.

It is heavy! Guessing about 1300 pounds or so but I guess that weight is needed for stability when it is full speed. By production standards, this machine is slow. I would estimate that it can do about a month or more of continuous hand-placed PCB's in a single day.

Couple of pics. It is currently sitting in my back yard while I make room in the garage.

[Corporate666]

Assuming your first picture is from the pick-up location, it appears he has another Quad 4C behind yours, in pieces.

Does this place strip/part/resell PnP machines?  Could use some parts

[ajb]

How is the control hardware arranged for these things?  Is it standard PC hardware, or something custom?

[rx8pilot]

The originals were DOS based PC hardware with custom interfaces and motor drivers. The one I got was converted to Windows which makes it better and worse all at the same time. Easier than DOS but more things to go wrong I think. Regrettably, I cannot mess with it until early net week. Should be fun to just look around and learn how it is put together. I need PCB's pretty quick so I am highly motivated to get going soon.

Assuming your first picture is from the pick-up location, it appears he has another Quad 4C behind yours, in pieces.

Does this place strip/part/resell PnP machines?  Could use some parts

They had two Quads one was a DOS unit with only a handful of feeders and mine was the Windows with 55 feeders. Both are operable. I think the thing in the BKG was some sort of medical machine. The pickup location was an commercial and medical electronics gold mine. I wish I could tear that stuff apart. They throw away TONs of stuff, amazing.

[Kjelt]

Regrettably, I cannot mess with it until early net week. Should be fun to just look around and learn how it is put together. I need PCB's pretty quick so I am highly motivated to get going soon.

Is that wise to use on very short time period a machine for finished products where you did not have enough time to figure it all out or assess the end quality? 
It could take you a lot more time than anticipated, do you have a fallback scenario in place?

[rx8pilot]

It could take you a lot more time than anticipated, do you have a fallback scenario in place?

Definitely, I have no idea what to expect. I only have hopes and will continue to keep backup plans in place until I get this machine verified.

Assuming your first picture is from the pick-up location, it appears he has another Quad 4C behind yours, in pieces.

I looked at the picture again, yes its another 4c. That one is all together and functional, checked out by a guy who runs them. DOS version. I wish there was a parts place in Los Angeles, but this place just happened upon them and really had no idea what they do.

Now that I have seen these things up close, I am VERY happy that I did not get into a TM240A or similar. With some patience, I got this machine that is vastly more capable for the same cost and what looks to be about a week of TLC to get it back to operational condition. Could be longer, but even if I spend a few $k, it will still be worth it IMHO.

[mikeselectricstuff]

Regrettably, I cannot mess with it until early net week. Should be fun to just look around and learn how it is put together. I need PCB's pretty quick so I am highly motivated to get going soon.

Is that wise to use on very short time period a machine for finished products where you did not have enough time to figure it all out or assess the end quality? 
It could take you a lot more time than anticipated, do you have a fallback scenario in place?

Having a definite goal is a good way to focus the mind on solving a problem...

[Corporate666]

Now that I have seen these things up close, I am VERY happy that I did not get into a TM240A or similar. With some patience, I got this machine that is vastly more capable for the same cost and what looks to be about a week of TLC to get it back to operational condition. Could be longer, but even if I spend a few $k, it will still be worth it IMHO.

Yep, "told ya so" )))

No offense to any other manufacturer or EEVBlogger or machine, but there is a WORLD of difference between a "real" PnP and anything else.  Wait until you get yours up and running... as a CNC/machinist guy like me, you will develop an appreciation for how it's all of the little things that make your machine so fit for purpose, and so difficult to replicate on a hobby scale.  Things like vision, on-the-fly centering, the feeders, motor control, SPEED and other things are simply difficult to achieve and are the hard parts of PnP.  X/Y/Z motion is easy and cheap.  Doing it at fast speeds, accurately and repeatably (within a few 0.001) is very hard and very expensive.

Once you get that machine up and running and see what it can do for the $5-6k you are into it for, you will appreciate why there is just no point in buying something like the TM240.  It's so crippled as to be effectively useless compared to a real machine.

I'm excited to see the post when you finally get her up and running!

[Kjelt]

Only real thing missing is the source code. In the future when some unsupported package comes along you might get stuck. That would be the greatest benefit from an open source open HW initiative, the community would keep it alive.

[miguelvp]

Only real thing missing is the source code. In the future when some unsupported package comes along you might get stuck. That would be the greatest benefit from an open source open HW initiative, the community would keep it alive.

And by community you mean a handful of developers or do you mean the rest that just enjoy the spoils and complain why things don't move faster?

[rx8pilot]

Only real thing missing is the source code. In the future when some unsupported package comes along you might get stuck. That would be the greatest benefit from an open source open HW initiative, the community would keep it alive.

The new software runs on Windows 7 so it will probably outlast the hardware. PPM is continuing to develop for the Quads so getting stuck seems rather unlikely. MANY people are still running the DOS versions without issue. Even my go to assembly house in southern California uses about 10 of them currently.

Keep in mind that my personal goal is to transition away from DIY processes and move into a true professional world. This is currently how I make a living so I have to be very careful about relying on community supported projects that have unpredictable results and timeframes.

And by community you mean a handful of developers or do you mean the rest that just enjoy the spoils and complain why things don't move faster?

So funny, so true.

[Kjelt]

Keep in mind that my personal goal is to transition away from DIY processes and move into a true professional world. This is currently how I make a living so I have to be very careful about relying on community supported projects that have unpredictable results and timeframes. 

Yes I understand that and the only reason you have support now is PPM. So that is probably why the Quads are still sold to companies.
A lot of unsupported (manufacturing company bankrupt or no support) P&P machines go very cheap because of that a normal company can not afford to buy it and run into unsupported problems.
Those end up with the wellwilling hobbieists.
A bigger company that has to make a living out of this will probably buy a brand new P&P machine with full support and write it off in 5 years or so.
Good thing for you there is still support for this machine 

[rx8pilot]

The only reason I am interested in the Quad is because of PPM and a few others. I think it will last a lot longer than 5 years because of that. My previous partner got a 'free' Panasonic PnP that was no longer supported so it was too expensive even though it was free with about 100 various feeders. In this case, I got a currently supported PnP with a solid feeder inventory for $5k. It was a delicately calculated move that did not happen overnight.

The Quads are very unique used machines because they have support. PPM is still selling fresh refurbs for $28k plus feeders. A full kit would be around $50k. For that money your options are very limited for a vision capable machine. Manncorp FVX maybe?

As a hobbiest, I am not sure if an old commercial machine would be much more than a project that needs a big air compressor, 3 phase power, and a TON of effort. I got very good at manual assembly and would probably never mess with PnP if I did not need the precision and volume.

[rx8pilot]

DAY 1:

Synopsis: Oh Shit.

I was out of town immediately after picking the machine up and today was the first day to take a close look. My goal was to get an understanding of the architecture of the machine since I could not even tell the front from the back. I also needed to get the thing in its final spot and get it off the pallet that it shipped on. Since it weighs well over a 1,000 pounds, it takes some thinking and patience to do it alone like I did. When I go to the machine shop tomorrow I will pick up the super level and get the table nice and flat.

The PC boots up but the application is stuck because of a licence problem - the previous owner did not pay. I have not yet seen anything move. PPM spent considerable amount of time on the phone today talking me through things and have not charged anything yet. I learned that if you try to game the license system it will brick the interface and you have to replace the PIC micro on the interface PCB. The are overnighting a new PIC for delivery tomorrow and giving me a temporary license to get the machine running. Cant wait to see that happen.

Right off the bat, I had to get dirty. The feet and the threads on the machine for leveling were damaged when it was removed from service. I had to insert Helicoils on the bottom of the machine for the new stainless leveler feet. I also removed the rack mount PC becuase the previous owner never mounted it properly and it was barely hanging on for dear life.



There was a serious layer of dirt. After speaking with PPM I learned that the machine had not been used since 2012. The owner did not pay for the perpetual license and machine will do nothing. A few years of dirt and MANY thousands of loose parts all over the machine. Passives of all types, SOIC, crystals littered the machine everywhere I looked. The operators REALLY did not give half a shit about this thing or the work they were doing. About 2 hours of cleaning and inspecting got it back to where it should be for production.

One pass with my hand.....


There were empty tubes, and about ten plastic cups filled with mis-picks. Many fell into every open crack available.


The enemy of neglect.....


The feeders all need service. Mainly the idler wheels that have disintegrated on may of them. I learned that these are the 12v 8pin models and apparently that is good news. 32 8mm, 16 12mm, 5 16mm, 5 24mm, 5 lane tube feeder, and two large matrix trays.
I hope that I can replace the idler myself. All the other functions seem to be good.


This does not look good....the wheels are falling apart.


The entire table is just a large and very heavy piece of steel with threaded holes. Anything and everything can be moved anywhere....its really cool. Very flexible.


The ball screws are very fast helix. It looks like the wrong grease was used but I should be able to get that worked out fairly easy. A fast ball screw requires a very fine encoder to get good position accuracy.


Linear encoders on this beast. These measure the actual position of the head as opposed to counting pulses of the motor encoder. Generally much more accurate and repeatable.


That is it for now. Hopefully I can get the machine to move around tomorrow when the PIC micro is installed. The feeders will take a while to get into production condition, but I can experiment and practice programming without the feeders.
This is awesome.

[ataradov]

So you buy a machine and it does nothing unless you pay more money? What's up with that?

[Falcon69]

Most industrial machines, like CNC Routers. Lasers, Pick And Place, etc. Require a yearly license to operate the software require to run it.  Most of these machines can only use the manufactures software to run them.

Where I used to work, we had a CNC router that we had to pay a yearly subscription fee to run it with the supplied software. If we wanted to use different software, we would have had to buy a complete separate Driver/computer system to work the machine, and then try and figure out how to connect all the servos, spindle, spindle attachments, etc. to the new system.  More pain than it is worth to just pay for the yearly fee.  It really does suck, pay $500,000 for a machine, than pay more money every year just to run it.

[rx8pilot]

This machine package was being liquidated for $5k. If I purchased the same thing refurbished from PPM, it would be about $55k or so. I assumed when I purchased that I would have to put something into it. The software license is perpetual for about $1500. Still not bad at all for what it will do.

Just the feeders alone are worth around $25k +

[Kjelt]

If I purchased the same thing refurbished from PPM, it would be about $55k or so.

You buy a brand new Mechatronika P&P M10V for €17k or MX70 for €25k but I guess this machine is better? 
Anyway I can not understand if someone would buy this machine for $55k, there are many more new and fully supported alternatives.
Ofcourse the money you spent is very nice amount if you get it up and running 

The software license is perpetual for about $1500.

So is the original manufacturer bankrupt and has PPM taken over their rights or why do you have to pay for software where there is no owner of? 

I assumed when I purchased that I would have to put something into it. Still not bad at all for what it will do.

Take a note of all the hours you will have to put into it , can add up.
BTW are you sure the phoneservice of PPM is free of charge, I mean it would be APITA if they sent you a bill for $75/hr afterwards 

[rx8pilot]

The machine is only $28k. Adding 55 various feeders is the rest of the cost. It can have up to 115 feeders plus trays which is substantial. Not sure if that is comparable to M10V or MX70.

The Quad line is owned by PPM, not just supported. Quad was purchased by Tyco and then sold to PPM. Tyco still has parts from what I understand. PPM has moved to Win7 now and I have the option of upgrading if Win XP gets too annoying.

[Kjelt]

The machine is only $28k. Adding 55 various feeders is the rest of the cost. It can have up to 115 feeders plus trays which is substantial. Not sure if that is comparable to M10V or MX70. 

Probably is, the prices I mentioned are starting prices so without many feeders probably which add considerably to the costs.
Can you get new wheels for your feeders or do you have to make them yourself?

PPM has moved to Win7 now and I have the option of upgrading if Win XP gets too annoying.

What I remember from my music workstation on XP is that you have to remove a lot of OS overhead stuff that can interfere with real time behaviour.
If tweaked correctly it can be very usefull.
I am not sure if in your case the windows PC does real time control or that there is another embedded platform in between that buffers all commands like with a CNC but probably because you have vision it is PC control only ? Just see how it goes, as long as you don't connect the pc to the internet (no security updates anymore for XP).

[mikeselectricstuff]

So you buy a machine and it does nothing unless you pay more money? What's up with that?

It cost a tenth of what it should have.

[mikeselectricstuff]

The software license is perpetual for about $1500.

So is the original manufacturer bankrupt and has PPM taken over their rights or why do you have to pay for software where there is no owner of? 

AIUI PPM have written new software, the original machine ran on DOS

[rx8pilot]

It's 'Sort Of' Alive!

I got the new PIC MCU (aka license security) installed and the machine moves! Big step forward for sure. The problem is that there is still some sort of issue with the vision controller card that is in the PC chassis. It is an OLD ISA bus controller that interfaces the PC with the image processor that is a separate sub system inside the machine. I cannot remember the last time I had to deal with an ISA bus and I never had to deal with it on a PC running XP. The PC running XP is a single board computer that plugs into a backplane with PCI and ISA expansion slots. I cannot tell if the problem is with XP needing some configuration tweak or if the old ISA hardware is broken. There are quite a few things that could be wrong from a shorted capacitor to a registry entry in XP.  Going back in time.

The only troubleshooting I have done on the card is measured the resistance of the power rail - 170 ohms. Seems low, but not ridiculous. No obvious signs of damage. The PCB is littered with old school 74xx logic, a pair of small FIFO CMOS memory (guessing these are buffers), and some mystery chips with 16bit hex address labels. Don't know where to start troubleshooting since I don't have any information at all about what this card does. A 'new' one is $200 to $500 depending on where I go to get it. PPM is convinced the board is dead, but that may be because they have a ton of them for sale at $500 each. I can't spend too much time on it so I may bite the bullet and buy the PPM one because they will actually support the installation troubleshooting if needed.

ICOS MVS922 Vision Controller. ICOS is now owned by KLA Tencor and I do not think they have any legacy docs that old.


UPDATE:
Still not sure what the card does in detail, but I did figure out that it has two RS485 ports that are communicating with the rest of the vision system. The PPM replacement is $500 plus my old card as a core exchange. I went for one that I found for $200 in unknown condition , but the seller said he will take returns if it is dead. At least then I will have one that can be used as a reference. Hoping it will work and hoping this is not the beginning of discovering a dead machine.

Most of the feeders need some TLC as mentioned before. I found all the parts and got prices, each feeder will range from $30 - $60 in parts but I do not need to do them all at once. I think I will go with a minimum for now to get some PCB's out the door and finish the rest over time. Total cost on the feeder refurb will be about $2500 US for 55 feeders, which is about the cost of 5 new feeders. I wish it was less but this is not yet a surprise. In my CNC machining days, I could blink my eyes and spend $2500.

[Corporate666]

Most industrial machines, like CNC Routers. Lasers, Pick And Place, etc. Require a yearly license to operate the software require to run it.  Most of these machines can only use the manufactures software to run them.

That's not my experience at all.. I've never seen an industrial machine that requires a yearly fee for the software.  Not to say there aren't some out there, but it has to be the exception to the rule, or specific to some industries.  I've got a lot of experience with PnP's, CNC milling machines and lathes, industrial lasers, routers and various robotics and I've never seen any that require an annual license.  Seems like highway robbery.

[Corporate666]

This machine package was being liquidated for $5k. If I purchased the same thing refurbished from PPM, it would be about $55k or so. I assumed when I purchased that I would have to put something into it. The software license is perpetual for about $1500. Still not bad at all for what it will do.

Just the feeders alone are worth around $25k +

They told you that machine is $55k new?   

They quoted me $28k for one and agreed to throw in whatever I wanted for feeders... or $23k for a refurbished/warranties machine with the DOS software, of $12k (or thereabouts) for a non-refurbished but guaranteed working model. 

[rx8pilot]

They quoted me $28k as well for the machine and I just added up the cost of all the feeders. How many feeders would they throw in for the $28k?

[rx8pilot]

With the machine at $28k, I added the asking price for the feeders from the 'reconditioned' prices and came up with $24,250. Total is $52k. They may throw in a few freebies if I walked in with a checkbook in hand.

[Corporate666]

They quoted me $28k as well for the machine and I just added up the cost of all the feeders. How many feeders would they throw in for the $28k?

When I talked to them about buying a machine a couple of years back, I think I had said I would need around 30 feeders and they said they would throw in whatever # I needed as part of the deal, I am positive I could have gotten a fair bit more.  I think PPM is starting to realize these machines aren't worth anywhere NEAR the prices they are asking for them.  At the end of the day, it's a 3600pph gantry PnP. There is an 8-head Philips Topaz on eBay for $12.5k asking (or best offer) - that is an 18,000CPH machine and Philips is a top brand that is still around.  There's a Juki 750 for $16k - an 11,000cph machine with feeders that I bet can be bought for $12k.   And a GSM Dual Beam for $19k that will do 20,000cph and I bet can be bought for $15k.   And when you get into the range PPM wants to charge for a windows machine with feeders, well there is a Philips Opal XII on Ebay for $48k that I bet could be bought for $40k.  That's a whole other class of machine - it will do micro BGA, 01005 parts at 15,000cph+.  So I think PPM's market for machines is probably shrinking real quick.  They have been in touch a few times to solicit me on the Windows upgrade.


As far as your machine - let me know you make out with the board.  Did you check your voltages on that card to be sure it's not getting fried by something upstream? 

[coppice]

They quoted me $28k as well for the machine and I just added up the cost of all the feeders. How many feeders would they throw in for the $28k?

When I talked to them about buying a machine a couple of years back, I think I had said I would need around 30 feeders and they said they would throw in whatever # I needed as part of the deal, I am positive I could have gotten a fair bit more.  I think PPM is starting to realize these machines aren't worth anywhere NEAR the prices they are asking for them.  At the end of the day, it's a 3600pph gantry PnP. There is an 8-head Philips Topaz on eBay for $12.5k asking (or best offer) - that is an 18,000CPH machine and Philips is a top brand that is still around.  There's a Juki 750 for $16k - an 11,000cph machine with feeders that I bet can be bought for $12k.   And a GSM Dual Beam for $19k that will do 20,000cph and I bet can be bought for $15k.   And when you get into the range PPM wants to charge for a windows machine with feeders, well there is a Philips Opal XII on Ebay for $48k that I bet could be bought for $40k.  That's a whole other class of machine - it will do micro BGA, 01005 parts at 15,000cph+.  So I think PPM's market for machines is probably shrinking real quick.  They have been in touch a few times to solicit me on the Windows upgrade.


As far as your machine - let me know you make out with the board.  Did you check your voltages on that card to be sure it's not getting fried by something upstream?

Its amazing how many models of PnP machine have been made. I wonder what the production volume has been like for most of these?

[rx8pilot]

Competition will do that. This industrial gear goes cheap when a company goes under or merges with another. I sure hope they are negotiable at PPM on the maintenance pieces and parts. The whole idea was to operate in a 'bargain' mode for a while and I will only operate this thing a few days per month. If I have to pour money into it and fiddle with it all the time, it will be frustrating for sure. If I can get a couple of years out of it with modest maintenance costs, I would be totally happy. I would not even slightly benefit from a 10,000 CPH machine or the maintenance costs associated with a higher end machine. One of the things I like about this model is that I can do so much of the maintenance. I would guess that I will place about <20k parts per month. In my previous manual efforts that would take about 55+ hours or local assembly house would be about $5,000. The problem that I have with assembly houses is that I have to document and kit the whole thing and provide a reference. After that, I have to wait a few days to get into their system followed by crossing my fingers they don't mess up - my fault or theirs dos not matter. For the foreseeable future, I will continue to do small batches which is deadly with contract manufacturing. At $5k per month in assembly costs, it does not take too long for this to make sense even working in the labor. I will never have the fingers crossed, can't sleep at night feeling with a batch run of a new design at the CM. I can setup and run one board and test it myself.

As I sit now, it looks like I will be able to make PCB's with about $2k in addition to what I originally paid - $7k total so far. If I can get them to soften up on the license ($1500) I would be pretty happy about that. All the other feeder parts and reel arms is about $2k additional. The whole license thing is because the previous owner never made the final payment so they never got a real license. They do not do annual licenses at PPM, this was an unusual case.

I did verify voltages, connections, and  IRQ's. I did some tracing on the PCB but gave up when I found a possible replacement for $200. If the 'new' one works, I will move on to the next big challenge.

Fingers still crossed that this is a good idea.

[rx8pilot]

Its amazing how many models of PnP machine have been made. I wonder what the production volume has been like for most of these?

It is truly nuts. We are surrounded by PCBs everywhere we go. They all traveled through a PnP somewhere.

[Corporate666]

They quoted me $28k as well for the machine and I just added up the cost of all the feeders. How many feeders would they throw in for the $28k?

When I talked to them about buying a machine a couple of years back, I think I had said I would need around 30 feeders and they said they would throw in whatever # I needed as part of the deal, I am positive I could have gotten a fair bit more.  I think PPM is starting to realize these machines aren't worth anywhere NEAR the prices they are asking for them.  At the end of the day, it's a 3600pph gantry PnP. There is an 8-head Philips Topaz on eBay for $12.5k asking (or best offer) - that is an 18,000CPH machine and Philips is a top brand that is still around.  There's a Juki 750 for $16k - an 11,000cph machine with feeders that I bet can be bought for $12k.   And a GSM Dual Beam for $19k that will do 20,000cph and I bet can be bought for $15k.   And when you get into the range PPM wants to charge for a windows machine with feeders, well there is a Philips Opal XII on Ebay for $48k that I bet could be bought for $40k.  That's a whole other class of machine - it will do micro BGA, 01005 parts at 15,000cph+.  So I think PPM's market for machines is probably shrinking real quick.  They have been in touch a few times to solicit me on the Windows upgrade.


As far as your machine - let me know you make out with the board.  Did you check your voltages on that card to be sure it's not getting fried by something upstream?

Its amazing how many models of PnP machine have been made. I wonder what the production volume has been like for most of these?

Even more amazing when you see all the ones that have come and gone!

I had two Dynapert machines - they used to be a huge company and just down the road from my shop.  I bought two giant chipshooters that did 18,000cph each.  Machines were huge and heavy.  I spent 2 years dicking with those machines and never did get them running right.  No parts, no service (company was out of business), rendered the machines useless

Had Celmacs 8700 machine (I think they are gone too?  Or maybe just in Japan).   I also had a Fuji IP3 (I think it was?).  When I called Fuji to ask a question, they asked what software I was running - when I told them, they told me there was no support, and practically hung up on me - wouldn't even talk about selling parts.

You will like your machine when it gets up and running.  I have a few of these machines so if you need me to check anything on mine for you, I can.

[Corporate666]

As I sit now, it looks like I will be able to make PCB's with about $2k in addition to what I originally paid - $7k total so far. If I can get them to soften up on the license ($1500) I would be pretty happy about that. All the other feeder parts and reel arms is about $2k additional. The whole license thing is because the previous owner never made the final payment so they never got a real license. They do not do annual licenses at PPM, this was an unusual case.

Who are you dealing with @ PPM?  Is it Sharon?  She's pretty cool - I think you should be able to do a bit better on the pricing, especially since it appears you walked into a problem that was not of your own making.

When you say "reel arms" - you mean the hanging brackets that support the reels on the side of the machine?  Did you get any with the machine?  It is just about the simplest thing I can imagine... if you have your CNC mill, you could crank a bunch of them out in no time out of 1/16" 6061 sheet double-sided taped to a piece of sacrificial stock.  If you didn't get any and want to try making some, I'd be happy to send one to you - I think I probably even have extras around somewhere (not sure but I can check).

[rx8pilot]

Who are you dealing with @ PPM?  Is it Sharon?  She's pretty cool - I think you should be able to do a bit better on the pricing, especially since it appears you walked into a problem that was not of your own making.

So far....Sharon, Tom, and Mike. All have been helpful, especially Mike. He has spent considerable effort talking me through the machine. I understand the electrical and PC systems much better. I am grateful that I started my career in computers and software. All my old school skills are becoming relevant for at least a moment. All the youngsters would be lost chasing a IRQ conflict on a non-plug-n-play ISA card. 

When you say "reel arms" - you mean the hanging brackets that support the reels on the side of the machine?  Did you get any with the machine?  It is just about the simplest thing I can imagine... if you have your CNC mill, you could crank a bunch of them out in no time out of 1/16" 6061 sheet double-sided taped to a piece of sacrificial stock.  If you didn't get any and want to try making some, I'd be happy to send one to you - I think I probably even have extras around somewhere (not sure but I can check).

Yes, those are the ones. I was thinking about doing that but I need to be careful that I don't spend a dollar to bend over a pick up a nickel. I may do better finding a waterjet company close by. That is an inexpensive process and I can send them a DXF file and be done. My guess is that they will be around $3-4 each instead of $20-30 each. That would save about $1,000 - hilarious.

[Kjelt]

The only troubleshooting I have done on the card is measured the resistance of the power rail - 170 ohms. Seems low, but not ridiculous. No obvious signs of damage. The PCB is littered with old school 74xx logic, a pair of small FIFO CMOS memory (guessing these are buffers), and some mystery chips with 16bit hex address labels

That is an easy and beauty to maintain this card. The "mystery chips" are probably Lattice Logic arrays (GAL/PAL) used for gluing the logic. You need to desolder the good ones once and read them in an eprom programmer (if not protected) store the contents and you can maintain this card for the next century if you can get the chips (not a big problem since they are standard series).
Nowadays these cards have one big blob on them, impossible to repair, so that is ofcourse a benefit of an old school P&P machine 

[rx8pilot]

Still working on the MVS922 (called the HS3L inside of the QUAD system). It is being discussed in a separate thread since I expected it to go on a tangent. I bought a used card that should be here in a few days and have a DigiKey order for most of the parts on the existing one. I have two out of three options in play so I am hopeful it will come alive and I will have a spare for the future.

In the meantime, I am reading the user manual. It is not very well written for someone that is not already familiar with the machine. It only covers programming, not the mechanics of the machine. I need to learn how the physical layout should be set up which is a little bit of guessing.

I did get the missing feeder parts called "reel arms" setup at a waterjet company down the street for $2.35 each. Plating and some plastic parts from McMaster Carr will have the cost up to about $4/ea. This is in contrast to the $20/ea for 8mm and $30/ea for 24mm for original parts. Total saved is about $1000 for 2 hours of effort. I feel like a lawyer for a moment.

I am also faking the rubber rollers that peel the cover tape off in the feeders. Another McMaster Carr solution will save about $900 or so. I would love to have the original parts and be done with it, but I am tight on cash and need to boot strap the operation a little longer.

@Corporate666 do you know anything about the IQ feeders. I believe I have the IQ feeders but not sure of the machine is setup to read them. From what I understand, they ID themselves wherever they are attached to the machine so you can program them offline and then the machine will know what it is and where it is located as soon as it is attached to the machine. Sounds cool, but not sure if I have that option.

[Corporate666]

@Corporate666 do you know anything about the IQ feeders. I believe I have the IQ feeders but not sure of the machine is setup to read them. From what I understand, they ID themselves wherever they are attached to the machine so you can program them offline and then the machine will know what it is and where it is located as soon as it is attached to the machine. Sounds cool, but not sure if I have that option.

My understanding on the IQ feeders is the same as yours... I think they were originally from the QSP2 machines and PPM adapted them to the 4C, but I may be incorrect on that.  Honestly, my previous 3 PnP's had parts management and I just shut it off, I found it a waste of time.  I can look at the reels and know what I have - and cheap passives like 0805 resistors, I buy 100,000 at a time and just buy another 100k when I am down to my last reel.  And more expensive stuff like SOIC MCU's and LED driver chips, I always know what my stock level is anyway.  Parts management is good if you are a contract assembly shop that keeps stuff reeled for customers, but for me it took more time than it saved.

I think the IQ feeders have blue LED's on them and the stock ones are red and green.  You can tell 12V feeders because they have 8 pogo pins sticking out the end and the 5V feeders have 3 pins.  If your feeders are IQ, I'd guess they would need something else in addition to the 5 pins on the end.  The rollers on mine are blue rubber, feels like neoprene.  I have some OEM feeders and some which were refurbed by PPM, and the 8mm ones all have blue rubber rollers.  12mm all have red rubber rollers.  What did you get @ McMaster to replace the rollers?  Let me know how it works out... I have some that could use refurbing. 

If your manual only covers programming, you are missing most of the manuals.  I have 2.. one starts off with setup (how to install feeders, programming feeders, teaching fiducials).  Then it goes into error codes and troubleshooting, then it goes into autoprogramming and such.  The back half of the manual is all the schematics and PCB layouts for all the boards in the machine.

Then I have another manual that goes into exhaustive detail on every little feature and option - both manuals are hundreds of pages long in a white 3-ring binder.

[rx8pilot]

If your manual only covers programming, you are missing most of the manuals.  I have 2.. one starts off with setup (how to install feeders, programming feeders, teaching fiducials).  Then it goes into error codes and troubleshooting, then it goes into autoprogramming and such.  The back half of the manual is all the schematics and PCB layouts for all the boards in the machine.

Then I have another manual that goes into exhaustive detail on every little feature and option - both manuals are hundreds of pages long in a white 3-ring binder.

When I picked the machine up, they pointed to a huge stack of paperwork so I thought there would be a ton of info. Turns out, 95% of it was proprietary manufacturing data that belonged to the customers of the assembly house that owned the machine. That is a good reason to have manufacturing in-house I guess.

DigiKey arrives tomorrow with components for the machine vision card and a 'new' card arrives on Saturday. I got the licensing worked out with PPM so I am all set with software. I have been cleaning and inspecting the feeders and no surprises so far. I do need to repair a cable harness that powers the feeder base but that is a quick job.

If the vision card works (just one of them), I should be able to start some test programming on Saturday. Today and tomorrow, I am having a guy help me get the PC backed up. Record all the BIOS settings and image the hard drive in case it craters. The last thing I want to do is a fresh install of XP so I will create multiple backups of a known good configuration.

Almost there.....

[rx8pilot]

Still struggling to get the vision system working. The 'new' used card arrived and I replaced most of the components on the original ISA card - nothing has changed.

I have, however, found other issues on the PC side that are troublesome. There was a PCI USB card in the computer that seemed to work ok the first time I saw the computer. After a few rounds of booting and connecting and disconnecting cables - that card had a problem. The PCI-PCI bridge was having a problem.....



This one....


I took the card out of the chassis, now the device no longer shows up on the list at all and the PC hangs on boot. It seems that it could be related to the PC being able to communicate with the IO bus. It the PCI is dead, I don't think the PCI to ISA bridge could work, rendering any ISA cards inoperable. I am not sure, but I think the ISA bus is controlled by the PCI bus through a bridge.

Also, strange problem with networking. When I plug an Ethernet cable in and it trys to connect - the whole PC freezes. For the moment, I will put this on hold until I can get the original drivers for the PC (Portwell http://www.portwell.com.tw/products/ROBO-8712EVG2A.html) and confirm all other settings and drivers from PPM. For now, I will spend some time inspecting the feeders.

[Corporate666]

On my DOS machine, the PC is only there to hold/edit data and to transmit it to the central controller of the PnP.  The PC just has an expansion card installed with a big fat centronix cable out the back which (on mine) goes into a serial port breakout board.  There are 2 serial cables going from that breakout board to the central controller, but during picking and placement, nothing is sent from the PC to the PnP machine, you can turn off the PC after you send the data.  And actually, you can even leave the PC off when you fire up the PnP, hit start, and it will continue placing boards with the last program that was loaded in.

In my machine, the lower left side of the cabinet (when facing the machine) contains the PC, and in the back is the air input and the transformer.  The lower right side front contains the main controller, vision cards, etc.  The rear is the servo drivers. 

So at least in my machine, there is not much in the way of connections between the PC and vision ISA card.  Is your vision card installed inside the PC itself? 

The PC shutting down or crashing with a card installed or not, and/or when trying to connect to the network isn't normal... if that was my PC I'd be starting with a fresh format and re-install of Windows, and expecting to have to replace the motherboard.

Let us know how you make out.

[rx8pilot]

I believe my machine is essentially the same. Some of the IO has been muxed into a USB now, but the vision and control system seem to be the same. As far as I can tell,  the PC simply sends a program to the control subsystems on the machine.

I will fiddle a little more this evening. I have a few emails out to PPM and the manufacturer of the SBC hardware. Hopefully some actionable info will arrive today.

[rx8pilot]

I send an email to PPM asking for any notes on the configuration of the PC - IRQ's, drivers, settings, etc, to confirm that everything is the way it should be before I buy a $500 card from them. The email response from PPM (Precision Placement Machines) is only pseudo encouraging. They say that "I’m still sure that the HS3L is defective. You don’t need to worry about the IRQ’s, or a driver for that board."

Bullshit.

There are IRQ settings and drivers required. There are possible conflicts or other PC side issues that could keep the card from working. In fact there are more things to go wrong on the PC side than the controller card itself. The solution is to pay them first and then hope it works. All I wanted to do was confirm a handful of setting to make sure the troubleshooting was being done properly. I have already paid $1,500 for a license that they claim the previous owner never paid for. Now that I have the software licensed, I can see how buggy it is. When I ask about any updates, they respond with updates will cost extra. I tried to clarify, that I don't want an UPGRADE, I want an UPDATE to make the thing I just paid for usable. I feel like $1,500 should get me software that does not crash when I push the first button.

I am also trying to explain to them that I need to be able to estimate the costs involved to get the machine running before I start spending. It has all the makings of being a runaway expense magnet. Oh....$1,500 for a license...oh $500 for a board.....oh $1,500 for software that doesn't crash.....oh the feeders need new urethane rollers, springs, etc for $3,000. I don't mind spending the money to do the right thing, but I will not be happy if I feel like I am being suckered into a black hole. I fully expected some challenges and so far it is not too bad. The responses I am getting from PPM indicate that they have no idea what is going on and/or they are deliberately keeping me uninformed so they can get more money. I will give them the benefit of the the doubt for a little bit, but I need intelligent and honest answers. As a business owner, I can fully appreciate that they need to make a profit from the efforts.
Lets see......

UPDATE:
The latest email explained the crashes I am seeing are because the software is unable to communicate with the vision system. When that gets fixed, the software should be good. Maybe all is well after all.

[rx8pilot]

Ok, I put what effort I could in the repair/diagnosis of the vision controller but it did not help. It does not make sense to pile on hours of labor when the card is only $500. I will order the part from PPM tomorrow and they have promised to take it back if it does not work and/or help me fix whatever else is not right. They have spent a number of hours on the phone with me so far, so I should not complain.

This gets the total cost to around $7k so far. My initial feeling is that if the total cost to get it to make PCBs is under $9k, it is still s great deal. If this fixes the vision system, $2k is enough to get the feeders up to speed.
We shall know on Wednesday I guess.

[Corporate666]

If the software is crashing because it can't communicate with the vision card correctly, then I'm sorry but that is total piss-poor software design.

Hopefully the new vision card solves the problem.  You're running into the issue I mentioned previously - they want to charge Juki/Fuji level support and service costs for a PnP machine that has a fraction of the value of those machines.  It doesn't really make the Windows upgrade look like a good on my machines.

Anyway, fingers crossed... keep us posted.

[Kjelt]

Ok, I put what effort I could in the repair/diagnosis of the vision controller but it did not help.

Did you ask them what a new guaranteed fully working PC with software etc. installed would cost?
Or how much if you sent them your PC that they bring it operational again?

[rx8pilot]

If the software is crashing because it can't communicate with the vision card correctly, then I'm sorry but that is total piss-poor software design.

I don't know about that. If the vision system is unavailable, the machine is generally useless so why bother making the software tolerant of that?

Did you ask them what a new guaranteed fully working PC with software etc. installed would cost?
Or how much if you sent them your PC that they bring it operational again?

Not yet....but that could be next. Re-installing XP is not my idea of fun. Hopefully it does not come to that. The PC is VERY simple and there is only a single application installed on it.
Calling now to get the new card on the way. I am not confident that it will work, but I will be very happy if it does. The trouble shooting process has forced me to inspect most all other elements of the machine, so I have a pretty good handle on how it works.

[mikeselectricstuff]

If the software is crashing because it can't communicate with the vision card correctly, then I'm sorry but that is total piss-poor software design.

I don't know about that. If the vision system is unavailable, the machine is generally useless so why bother making the software tolerant of that?

Not ideal, but it would be nice to at least give a meaningful error message

[coppice]

If the software is crashing because it can't communicate with the vision card correctly, then I'm sorry but that is total piss-poor software design.

I don't know about that. If the vision system is unavailable, the machine is generally useless so why bother making the software tolerant of that?

You don't think diagnostics are important? If the software can't communicate properly there is no excuse for it not to detect that condition and give a reasonable report as to what went wrong.

[mikeselectricstuff]

If the software is crashing because it can't communicate with the vision card correctly, then I'm sorry but that is total piss-poor software design.

I don't know about that. If the vision system is unavailable, the machine is generally useless so why bother making the software tolerant of that?

You don't think diagnostics are important? If the software can't communicate properly there is no excuse for it not to detect that condition and give a reasonable report as to what went wrong.

Even if it's not practical to detect without crashing, which may be the case in some circumstances,  at least something like "Attempting to access video card" beforehand would give a good heads-up.

[rx8pilot]

You don't think diagnostics are important? If the software can't communicate properly there is no excuse for it not to detect that condition and give a reasonable report as to what went wrong.

It does clearly show the error during initialization in a message window. The error does not, however, stop you from trying to use the machine. I can jog the axis around and do a few other things. If I push a button that needs the vision resource, it will crash. Is that ideal? no. Is it typical in ultra-low volume specialized software that is relatively cheap? Yes. $1,500 for the software does not go very far towards covering regular monthly expenses. Not to mention, even a rough software package costs huge money and there are only a small number of Quads to be converted.

The more I see of the machine, the more the support costs make sense. I would love for it to be cheap or free of course. Once this machine is up and running, it will be critical that a stocking source of parts and service is only a phone call away even if I don't need them very often. Machines always seem to break just before the biggest job of the year and I will not have time to fiddle with it. If PPM goes away, this Quad machine will plummet in value just like the Panasonic machine my former partner got for "free". The Panasonic was totally unsupported and therefore FAR more expensive than this machine. The business end of this is delicate to be sure. I don't have the knowledge or experience, so I have to buy that from someone.

[Kjelt]

Another thought
Might be a good idea to clone the original hdd to a new one.
I hope it has a Sata drive not an Ide otherwise go shopping for some replacements.

[rx8pilot]

Another thought
Might be a good idea to clone the original hdd to a new one.
I hope it has a Sata drive not an Ide otherwise go shopping for some replacements.

Absolutely! It is an IDE bus so I can go with IDE to SATA adapter or better yet I found an IDE SSD. One of my friends is going to supervise that effort. The plan is to make a clone to the SSD and and make an image that can be backed up to our SAN. Someday, it will surely save the day. I am considering what spares I should keep on hand, but after speaking to PPM, they have pretty much every single part I would ever need on the shelf. They charge a premium, but it is less than me having to do it on my own. With my luck, I would have various spares on hand, but the thing I don't have is what will break.

If I push a button that needs the vision resource, it will crash. Is that ideal? no. Is it typical in ultra-low volume specialized software that is relatively cheap? Yes.

I disagree. It is almost always a sign that somebody didn't know what he was doing (or he didn't care). Crashing in such common scenarios is a big indication of poor quality software.

I don't yet have a well formed opinion because I have not used it, but I am ready for a few negative surprises from the software. I would say the chances of this being a super performing software package are rather low. All I need is to figure out the work-arounds and get a small number of programs up and running. For CNC, programming, I have spent $12k to $16k each on software licenses from two companies only to discover the software was rotten to the core. The solution, keep paying $2k per year for updates that don't fix the problems, but just add new features that I don't need but marketing needs to sell more licenses. This is on top of the CNC hardware, tools, SolidWorks, PC hardware, and more. For $1,500, I am not expecting much at all.

The options are to revert back to MS-DOS version, or write my own software for maybe $500k or so.

[miguelvp]

Accurate
Reliant
Robust
Modular
Performance

Pick just two.

They probably put more effort in being accurate and robust with all the pieces in place than anything else.

Since it needs the camera to operate properly I don't see a problem with the program crashing, I rather it do that than keep on working with all kinds of placement errors.

[rx8pilot]

Shes ALIVE!

The problem with the ISA vision card and the networking appear to be related. I re-seated the SBC and the RAM and it all came alive, no issues!

The ISA card 'repair' attempts were documented here https://www.eevblog.com/forum/repair/finding-fault-in-early-90's-isa-pc-card/45/ Like I said in theat thread, I am so embarrassed I did not try that earlier in the process - I bet someone told me to do it along the way. Kinda scared to go back a re-read the posts.

Anyway....we move forward from here. Now I can actually program the machine and figure out how to make PCB's with it.

Super excited about this change in fortune.

[Kjelt]

Great 

[rx8pilot]

OK, now that the basics of the machine are looking good I can focus on the feeders.....

At first glance, the feeders and the various serviceable parts looks inexpensive. When I add it all up for 55 feeders, it's a few thousand dollars. I also learned that the joystick remote pendant is damaged. That is the hand held box used to jog the head around while programming. It could be considered a convenience item since the job can technically be done with out it. From what I have seen so far, it could be a pain in the butt without it. I took a look at it and it appears the handle fell off which allowed the springs to come out. Looked like an easy fix, I put it back together but the joystick module does not work - it remains electrically 'centered' regardless of position. DigiKey and a few others have it listed on the their sites but there is no stock. 2 month lead time.

http://www.digikey.com/product-detail/en/9SA50RE6500/679-1414-ND/1280480

No biggie, I call PPM to see if they have any and found that they do not sell the parts, only the whole hand held unit that is $1,800. Ouch. This is a budget operation and that repair is going to have to need a DIY solution of some sort. The joystick part seems to be ultrasonically welded plastic. Not sure if a hack is possible. For now I am looking for a newer version of the same thing that I can just drop in. I don't mind spending $200, just not $1,800.

For you viewing pleasure. I present the hand-held pendant that controls a PPM Quad IVC. For those familiar with these machines, it is only used on the modified Quads with the newer Windows configuration.



Now.....this is what you get for $1,800. I don't want to bitch too much because PPM needs to stay in business, but this is not what I expect from $1,800. A drilled project box and speaker wire zip cord? The Joystick itself is about $200, the rest of it amounts to a few dollars. Trust me, I understand what making small batch specialty electronics is all about - it is what I do. For $1,400 I sell a product with a CNC machined modular case, 7 multi-layer PCBs, graphic LCD and a bunch of Swiss made connectors and doing very well at it.


Hopefully I can find a compatible joystick, otherwise I will rig something up until Digi Key can get me a real one.

[Falcon69]

rx8pilot

Some folks over on CNCZone.com are using PS3, X-Box, etc. controllers to do the jogging and movement on some of their DIY CNC machines.  Might want to check it out. May be a cheap alternative to replacing that Joystick.  Just do a search on their site, and You'll see alot of mods for the joysticks.

[LabSpokane]

Wow. Crimp ends on lamp cord. That just oozes class.

As for the joystick, if that's just a pair of pots on a stick, you should be able source a short term replacement fairly easily.

[miguelvp]

If it helps here is the datasheet for the stick

http://datasheet.elcodis.com/pdf/19/31/193131/26.pdf

[rx8pilot]

The APEM joysticks appear to be rather good quality. In the end, it is just two pots but this model uses an inductive contact-less configuration that seems to be for longevity and predictable response. Its an active circuit. Digi-Key has the Hall Effect versions in stock but I am not sure how they are different. The original has a 'center' tap reference that I am guessing the PIC18F controller uses. The Hall Effect sticks only have the X/Y wiper outputs. I guess I could feed the center reference from a hand wired voltage divider and accomplish the same thing.

If I can use the Hall Effect versions I would be golden. Cheap and easy without making a big project out of it. The PCB runs from 5V so it would be compatible with these alternates.
http://www.apem.com/files/apem/brochures/Joysticks_2015/APEM-HF-Fingertip.pdf
http://www.digikey.com/product-detail/en/HFX11S10/679-2262-ND/2063253

[rx8pilot]

Me thinks I have located the problem, but not sure there is a practical fix. Great learning opportunity though.

I took the stick apart which was easier than expected. The bottom of the stick is an inductor that moves around 4 fixed inductors. The mechanics are pretty slick and high-quality. The problem I see is the resistance of the movable inductor is 5M - rather high I would say. Under my microscope, it looks like the magnet wire may be broken, but in a place that I probably cannot get to without damaging something else.





UPDATE:
I figured what the heck, may as well try....
I removed the coil carefully by cutting it under the microscope. It looks like 1 million gauge wire, unbelievably delicate. Even under magnification it's small. After a few deep breaths I sacrificed one turn to get enough wire to make the connection. Did a quick check and we are down to about 35 ohms which seems about right. Put a tiny dab of glue to hold it in place while I test it out. If it works, I will glue it properly. Unless the circuitry is bad, it should be all good to go.

[Corporate666]

I think you said you bit the bullet and ordered the vision card from PPM which was coming Wednesday - then the issue turned out to be a bad connection.  So just curious - did you send the $500 one from PPM back, and did they take it without hassle?

I've fixed lots of little parts on industrial machinery by purchasing parts from eBay.  There's several APEM Joysticks on eBay right now, including one that's an inductive 2-axis model.  Not the exact same PN, but looks very similar.  There are also lots of other inductive joysticks on eBay for much less, like this:

http://www.ebay.com/itm/Inductive-Joystick-Replacement-for-Everest-Jennings-Z60-Z51i-Z45-/141093173758?pt=LH_DefaultDomain_0&hash=item20d9cef9fe

Just something to think about.

I'm sure most folks understand that companies have to make a profit, but IMO that's no reason to do stuff half-assed, like using freakin' lamp cord in an industrial housing.  That's just hack level crap.  A nicely made anodized/laser engraved CNC housing wouldn't be too expensive (especially compared to a custom silkscreen printed one like they are using now).  I think looking inside that enclosure sums up the overall approach to the Windows upgrade.

[Kjelt]

I'm sure most folks understand that companies have to make a profit, but IMO that's no reason to do stuff half-assed, like using freakin' lamp cord in an industrial housing.  That's just hack level crap. 

+1 I am a hobbieist and have 9 different colors cable in 0,75mm²,0,5mm² and 0,25mm² on a 100m and 250m rolls in stock. 

[rx8pilot]

Yes, the vision card is going back. I was not convinced that it was the card so I had the conversation about returns up front. No hassles at all from PPM. They have been quite pleasant through this effort. I will be trading the credit for various feeder related parts so they are still getting some biz out of it.

The joystick is confirmed working, rather happy about that . It took longer to post here about it than to actually do the fix itself. Confident it will be a lasting repair. No $1800. No $215 plus 8 week lead time. No substitute part that could be a project to work out. Done.

Now, all errors are cleared and the machine comes up just perfect. It appears the calibration is good as well, although I need to confirm that to be sure. I am running out of excuses fairly quickly - closer to making a PCB!

[rx8pilot]

As I dig through the feeders I am realizing how important it will be to splice the tapes since the leaders are needed to thread a feeder.

I looked at http://www.tapesplice.com/tapesplicerandtoolkit.htm but the starter kit is $600. Seems like it works well, but I was wondering if there is a less expensive way to get started. I have lots of high value parts in short strips that I have been using in my manual tape holders.

Anyone have any SMT tape splicing tips?

[mikeselectricstuff]

As I dig through the feeders I am realizing how important it will be to splice the tapes since the leaders are needed to thread a feeder.

I looked at http://www.tapesplice.com/tapesplicerandtoolkit.htm but the starter kit is $600. Seems like it works well, but I was wondering if there is a less expensive way to get started. I have lots of high value parts in short strips that I have been using in my manual tape holders.

Anyone have any SMT tape splicing tips?

Does it need both sides of the tape as leader ?
Remember Digikey/Mouser do a reeling service to add leaders to cut tape, subject to a MOQ
I'd think you could probably do short tapes from a fixed feeder like a tray.
 

[rx8pilot]

Digi Key does the splicing as you mentioned for about $7 per reel. I would not take long to justify the splicer I guess. My current set of PCB's have a total of about $175 in components. That value keeps me from buying in full reels at this time. About 20 components would be subject to the "Digi-Reel" fee for a month of production - so about $140 in reel fees. That is not too bad actually since it arrives ready to rock.

I have about 100-150 cut strips already delivered that would be nice to use. I can buy or build a cut tape feeder as well, but that would probably be just as much effort as splicing and loading into a regular feeder. I seems like a great thing to have, but hoping to find a 'beginner' priced tool for now.

It needs the carrier tape and the cover tape to thread - seems like about 6" or so is about right.

[mikeselectricstuff]

That's one downside of using "proper" machines - they assume you use everything by the reel...
Though I've seen some short-tape feeders on Mydata - not sure if these are standard or a special variant
 

[rx8pilot]

Yes, "proper"  machine can mean "proper" prices. This Quad has a vibratory feeder for parts in a tube. The tops are interchangeable for different sizes. Of course mine was delivered with PLCC version and I need SOIC-8. I got  prices of $995 from two sources. It is a flat piece of aluminum with 8 shallow slots milled. EXTREMELY basic manual machining job for $995?

All the parts are expensive. The reason this is still a good idea is because I have been able to do the work. If a technician came out to get thing up and running, it would not likely be worth it. Or, at least I could not have of afforded it.

I will feel better when PCB's are being made. For now, it still seems like a great way to 'graduate' from hand built prototypes and small batches to the next level. I can tip-toe into higher sophistication and higher volume.

[Corporate666]

How short are your strips?  You can program the machine to do picks in an X/Y (or just X or just Y) grid which is how the machine handles waffle trays.  And since it has vision correction, all you really need to worry about is getting a clean pick on the part.

If it were me, I'd take a scrap of aluminum and just machine several slots in it, mount that on the machine and pick from it.  Or if you don't have that option, you could just use a ruler and line the cut strips up on a piece of flat aluminum and just use double sided tape to stick them down.

Moving forward, I like the Digikey reeling service and I use it for high value parts that I dont want to buy full reels of.  There are also always local companies that will do reeling, even of loose components.

[Corporate666]

Oh and as far as splicing...  I would never pay $600 for a tool that didn't do the job almost completely automated.  Reasons are..

1) I rarely have tape feeding problems, but it's always with spliced reels.   And I've had it from numerous sources, so it's not just one supplier - it's just that spliced tapes don't feed as reliably as intact tapes.

2) Splicing reels myself is super easy.  A rubber cutting mat, sharp scissors, some kapton tape and about a minute of your time produces just as good results as "pro splicing".  You can get Kapton tape at Uline

http://www.uline.com/Product/Detail/S-14532/High-Temperature-Tapes/Kapton-Tape-1-Mil-1-8-x-36-yds

and just a 2" strip of Kapton on the top and bottom side of the cut strip is fine.  I haven't noticed any difference in reliability between Digikey/Mouser spliced reels and the ones I do myself.


You can buy blank tape with no components (but with the plastic top cover), but most reels have blank tape at the end of the reel also.  And all reels have some at the start... the Quad 4C feeders need maybe 6 inches of blank reel, and most components will have a fair bit more than that.. just trim it off and there is your lifelong supply of blank-but-covered tape. 

[mikeselectricstuff]

Just found this video on the PPM'd IVC
http://lowpowerlab.com/blog/2014/05/13/ppm-visit-and-quad-ivc4000c-pick-and-place-demo/

[PA0PBZ]

Just found this video on the PPM'd IVC

It has the Windows 7 logo so it must be cool 

[rx8pilot]

Oh and as far as splicing...  I would never pay $600 for a tool that didn't do the job almost completely automated. 

Well, to be fair, the tool is half that. The rest of the cost is the consumables ranging from 8mm to 56mm. Seems simple enough to deal with flat paper tapes but the deep pocketed plastic tapes for bigger components like inductors and push buttons seem harder. A website called http://www.nozzles4smt.com/ has a cut tape carrier that is adjustable to use various width tapes. Looks promising for my kind of work. They also have other consumables for Quads - nozzles, rollers, etc.

Just found this video on the PPM'd IVC
http://lowpowerlab.com/blog/2014/05/13/ppm-visit-and-quad-ivc4000c-pick-and-place-demo/

I saw that video shortly before finding this machine. His operation seems similar to what I am doing. Wonder how much he paid for the package. It looks like after my machine is all dialed in and ready to go, it will be around $8k plus about 40-50 hours of labor working the various problems out. The list for everything being over $50k (maybe PPM would have given me a 'deal') seems like it is a pretty good deal so far.

The feeders have been all inspected. I made a log of them to keep track of various issues. In general, they all seem to be in working condition with the exception of the part called a Peel Roller. This is a urethane roller that pulls the cover tape off. Very simple, but for some reason, MANY of them melted like candle wax. Looks like they were not chemically stable from the beginning. I have never seen such a thing from urethane.

This is what I saw.....check out the goo.



Some were even worse....not easy to clean up the mess.



I will post more later....

[rx8pilot]

Ok, I had to go pick up the kids.

So I now have a list of the replacement parts I need to get the feeders working. I was pleasantly surprised that the urethane rollers were really the only problems. The circuit boards, mechanics, gears, etc were all working well. Many of them had components rattling around inside. This was mainly the 8mm ones that deal with the small parts. Some had tapes tangled up inside as well. All in, it was about 10 hours to go through all the feeders, log, inspects and test. It was mind numbing, but necessary. I wanted to send them to PPM or anyone else that services feeders, but it was out of my budget. PPM charges $75 to inspect a feeder, another guy in southern California offered a refurb service for $125/ea. With the number of feeders I have it would cost $4,350 to have them all looked at - not including the replacement parts. It would say that the 10 hours I spent learning how they work and cleaning was well worth it.

I learned a lot about how they work and overall I would give the design a B+ grade. These feeders appear to have been abused pretty bad, yet they still work just fine. That is worth a good grade. They also have the mechanics exposed too much so dust and parts can fall into the PCB and gears which of course is bad. I cannot quite understand the melting roller problem. I think the previous owner may have sourced these parts from some shady place. Many of the feeders had hardware store mods to the rollers using grommets or O-rings to rig up the feeders. Some had scotch tape around the rollers that were going bad. I will be maintaining these feeders and the machine far more meticulously than that. Not because I am OCD, but because I need this machine to be reliable and a clean and well maintained machine is the beginning of having a reliable machine. That lesson was learned in my CNC machining days.

So, when the feeder parts show up this thing should be ready to roll. I am now focusing on how to program it. There are a few videos that PPM has made and they are helpful. I have the challenge of learning the software without knowing much about PnP which will no doubt make things more challenging. When I decided to get into CNC machining, I first got a CNC machine and then started looking for the power switch.


Here is a pile of melted rollers from the feeders. About 20% of them looked like this.


This is the setup to clean the feeders. Lots of coffee as seen on the left.


The feeder with the cover removed reveals the PCB and the gear train. The belt is a consumable, and sheds all over the place.


Two of the feeders were a little different and the PCB said Samsung Aerospace. Didn't see that coming. I thougt Samsung made competitive machines.

[rx8pilot]

The pick and place machine is now up and calibrated. Technically this means that it is ready for programming. The problem is that I have no idea what I am doing.

I get the concepts - pickup part, place part, pickup part, place part. There are, as I have discovered, quite a few details that need attention. The programming manual is 390 pages and the 'introduction' to programming video is 4 hours and 20 minutes. Needless to say, I have a bit to learn.

From the overview of the materials, it seems that a lot of it is targeted to assembly houses that are constantly changing the whole machine. In my case, I will get a setup dialed in and rarely change anything. The way it is designed is very flexible allowing anything to be located where ever I want it. While I appreciate that flexibility, it comes at a price. When things move, they have to be calibrated and verified. It is not difficult, but it takes enough time that I will try to avoid changes if I can.

The software seems to be stable so far, but it reminds me of the learning curve experience with Eagle CAD. It is clunky and not intuitive. It does not really go with typical Windows conventions. For example, there are buttons with fly out menus. At first glance, the menus all look like buttons and there are buttons that are just buttons. You have to just know which buttons are really a menu. Also, there is not really a logical flow that I can tell yet. Bear in mind, I am a PnP newbie as well so maybe it will be more apparent when I am more versed in the task. I have yet to find any kind of a work flow to help guide a new board. The videos and the printed documentation do not seem to cover the task as a flow - they just go from menu to menu describing what functions are available.

I love the software flexibility that I see so far. There are many defaults that the user can define, but almost everything can be overridden. This is  very similar to what I need in CNC machining software. The defaults get me in the ballpark very quickly and while still allowing me to finely tune every detail if needed. For example, you can control the vacuum pressure on each and every part as well as the delay time to allow the vacuum to build. Handy for heavy inductors, but not needed of 0402 passives. you can easily control speed as well to avoid having a heavy or wobbly part from slipping after it has been through the vision system which would result in a bad place.

Not sure how other commercial machines deal with alignment, but this one has two methods. The 'Quad Align' is a system that is attached to the placement head and deals with alignment in while in motion. It is essentially, a line scan camera on one side and a light source on the other side. It looks at the shadow of the part while spinning around. The Quad Align system can detect X/Y/T and Z offsets relatively well. It is fast and does its whole routine while traveling from the feeder to the PCB. For fine pitch parts and BGA's the you can choose the upward facing camera to get really precise placements. This is also where the flexibility of the software shines. You do not have to choose one or the other - you have have an attempt with Quad Align, and if it fails the system can do a camera alignment. You can also define what to do if it fails alignment - try again, try x times then stop, or just stop.

This level of detail is why I wanted this machine over something like a TM240a or one of the other low-cost options. This machine can deal with just about any board and any part with ease. As I get past the weirdness of the software design, I feel like it will be quick and easy. I did write a script to get my data out of EAGLE in a way that perfectly matches what the machine needs - including nozzles and package types. This will slip into the machine defaults nicely and get programs up and running fast (I hope)

I have no reference for the DOS versions so I cannot say if any features in my Windows upgraded version are new or any better.

All the feeder parts arrive tomorrow, so I should have 100% of my feeders ready to go.

[Corporate666]

Glad it's (nearly) working.

You will be jumping for joy when you first see the machine working, and will instantly realize why a "real" PnP is such a difficult thing to build.. .it's all the little things that make it hard.

On your feeders, I bet someone used something like toluene or MEK to clean them... I've never seen urethane just melt like that, but if it's not cross-linked, it will just absorb the chemical and over time, turn to crap.  Did you end up buying rollers @ McMcaster?  What specific parts did you buy, if I may ask?

As far as programming, a couple of tips... when you program your pickups, you want to get to where the nozzle is just about touching your part.  The easy way to tell is to get it close and turn on and off the vacuum.  As long as you see the part jump up a little, the nozzle isn't touching.  Jog the Z down a few thou until the part doesn't move when turning on/off the vacuum... then jog down maybe 5 or 10 thou more for a good pickup location.

With placements, take a blank PCB and apply thin double sided tape.  Jog the part down slowly until you just see the Z spring/bellows start to compress, and set that as your placement Z level.

I've never had to adjust the pickup pressure - but you do want to be sure you are using the right nozzle.  For larger/heavier parts, use the nozzles with the rubber grommets that seal against the part (SOIC's, inductors, etc).  And make sure you have the offset between the nozzle position and camera position set dead-nuts accurately.  This is critical when you are using the camera and crosshairs to set your X/Y location for pickups or placements.  Take the time to get it perfect.

Rather than trying to get a whole board done first, take a blank PCB and use the double sided tape and just try to write a program to place one part - then a few parts, then a few different parts.  Then figure out the repeats... component repeats and PCB repeats so you can do groups of components and panels of PCB's.

Once you have that down, you pretty much have it all.  I don't bother failing over from Quad-Align to the upwards camera... Quad Align is 99.9% accurate and is used for all passives and SOIC's.   If you get into bigger stuff like QFN's or TQFP's, I'd use the upwards camera 100% of the time - it's slow compared to Quad-Align but accuracy is more important for higher value parts.

[Kjelt]

With placements, take a blank PCB and apply thin double sided tape.  Jog the part down slowly until you just see the Z spring/bellows start to compress, and set that as your placement Z level.

How accurate is the baseplate of the machine where the pcb is mounted in the Z axis,  compared to the head.
Of differently stated, do you have to do this calibration for all parts or just some at the edges and in the middle of the pcb?

[mikeselectricstuff]

With placements, take a blank PCB and apply thin double sided tape.  Jog the part down slowly until you just see the Z spring/bellows start to compress, and set that as your placement Z level.

How accurate is the baseplate of the machine where the pcb is mounted in the Z axis,  compared to the head.
Of differently stated, do you have to do this calibration for all parts or just some at the edges and in the middle of the pcb?

Don't know about the quad,  but once the PCB height has been measured, it should be able to do this automatically from the component height data in the part library ( which is also used to determine the focus height for the fixed camera). 
It would be a major ball-ache to have to do this for every part.

[Corporate666]

With placements, take a blank PCB and apply thin double sided tape.  Jog the part down slowly until you just see the Z spring/bellows start to compress, and set that as your placement Z level.

How accurate is the baseplate of the machine where the pcb is mounted in the Z axis,  compared to the head.
Of differently stated, do you have to do this calibration for all parts or just some at the edges and in the middle of the pcb?

I haven't measured it with a dial indicator, but one of the critical adjustments of the machine is coplanarity of the X/Y axes in relation to the base plate.  From seeing how my machine reacts, it can't be out more than a few thousandths of an inch, if that.

PCB's have a fair amount of spring in them, especially larger boards or panels towards the center.  The Z-axis of the Quad has a spring loaded bellows system (basically just two rods that slide inside one another with a spring between them).  You want to program your parts such that during placement, the Z spring will slightly compress - this ensures that the part is always being pushed down into the paste regardless of variations in PCB thickness or coplanarity offset errors.

As far as setting the parts, you set pickups and placements.  A pickup is essentially 4 datapoints - part number, X, Y and Z.  A placement is 3 datapoints - X Y and Z.  So you can set the placement height for every part, and in general when programming a board you need to set different Z values for something thin like an 0803 vs something thick like an SOD-123.  However you normally just copy/paste the Z value from prior placements - I don't think you would bother to change it unless there was specific reason.

EDIT: Something to keep in mind is that with the standard conveyor system, the PCB is on rails and is pushed upwards to lock it in place - meaning the top surface is always in the same plane, whether the PCB is 1.6mm, 0.8mm or whatever.

With a fixture system where the PCB is placed down on a flat location, variations in Z height would need to be accomodated.  However, there are "mod codes" (aka - settings) for global X/Y/Z offsets, per-pickup X/Y/Z offsets, per placement X/Y/Z offsets, per program X/Y/Z offsets - and you can change the program itself.

[rx8pilot]

You will be jumping for joy when you first see the machine working.....

I will go straight through the roof with excitement! This has been a LONG time coming. For me, its not just getting the PnP, it is the beginning of having a complete electronics design business. I am managing to make a living even while hand assembling all the PCB's. Moving into assembly automation will give me a LOT more time to focus on designs and continuing education.

On your feeders, I bet someone used something like toluene or MEK to clean them... I've never seen urethane just melt like that, but if it's not cross-linked, it will just absorb the chemical and over time, turn to crap.  Did you end up buying rollers @ McMcaster?  What specific parts did you buy, if I may ask?

I will never know exactly how they came to be melted globs for sure. It could have been that some random urethane house made them and the chemistry was never correct and/or the end user as you say used some cleaner that un-linked the polymers somehow? Definitely not an area of knowledge for me. As I spent considerable amounts of time cleaning up the mess, I decided to buy the 'real thing' from PPM. I have saved so much money on this project by learning and doing everything on my own, that I should just buy the right parts instead of trying to fake it.

As far as programming, a couple of tips... when you program your pickups, you want to get to where the nozzle is just about touching your part.  The easy way to tell is to get it close and turn on and off the vacuum.  As long as you see the part jump up a little, the nozzle isn't touching.  Jog the Z down a few thou until the part doesn't move when turning on/off the vacuum... then jog down maybe 5 or 10 thou more for a good pickup location.

With placements, take a blank PCB and apply thin double sided tape.  Jog the part down slowly until you just see the Z spring/bellows start to compress, and set that as your placement Z level.

I've never had to adjust the pickup pressure - but you do want to be sure you are using the right nozzle.  For larger/heavier parts, use the nozzles with the rubber grommets that seal against the part (SOIC's, inductors, etc).  And make sure you have the offset between the nozzle position and camera position set dead-nuts accurately.  This is critical when you are using the camera and crosshairs to set your X/Y location for pickups or placements.  Take the time to get it perfect.

Rather than trying to get a whole board done first, take a blank PCB and use the double sided tape and just try to write a program to place one part - then a few parts, then a few different parts.  Then figure out the repeats... component repeats and PCB repeats so you can do groups of components and panels of PCB's.

Once you have that down, you pretty much have it all.  I don't bother failing over from Quad-Align to the upwards camera... Quad Align is 99.9% accurate and is used for all passives and SOIC's.   If you get into bigger stuff like QFN's or TQFP's, I'd use the upwards camera 100% of the time - it's slow compared to Quad-Align but accuracy is more important for higher value parts.

Thank you! The printed docs and video don't even come close to offering tips like that. Taking baby steps will keep me from destroying things and save time. Resistors are nearly free.

Don't know about the quad,  but once the PCB height has been measured, it should be able to do this automatically from the component height data in the part library ( which is also used to determine the focus height for the fixed camera). 
It would be a major ball-ache to have to do this for every part.

The Windows software has a library of part packages with height defined. The script I wrote to get the placement data out of Eagle matches the Eagle package with the PnP library, so in theory, the Z height is pre-defined for each and every part before I push a single button. With a new part, I would have to create the library entry for it. This also matches the feeder type required for that part. If all works as designed, I should be able to import data and the PPM software should be able to assign a feeder, a nozzle,  and know the Z level. All I need to do is get the PCB physically in the machine. In practice, I am expecting that tweaking will required. Hopefully only a little though.

For the PCB flatness, as Corporate666 says, it will flex. The machine itself is very flat and predictable but the PCB is less so. The Zaxis is on a spring so it can deal with variation in actual Z-levels. The 'mod codes' have been replaced with 'offsets' in the Windows system - but it sounds like the same thing. You can make global changes, changes to a group, changes to a package, or changes to an individual pickup.

[Kjelt]

Thanks all for the answers and insights. 

[mikeselectricstuff]

The Windows software has a library of part packages with height defined. The script I wrote to get the placement data out of Eagle matches the Eagle package with the PnP library, so in theory, the Z height is pre-defined for each and every part before I push a single button.

and of course you want to set things up so parts in your PCB libraries match the zero rotation of parts in the feeders, so things end up the right way round by default.

 Something I would advise is to use different part names for different types of parts in the same package, e.g. 0805 resistors, caps, LEDs should use different package names in your PCB library, with corresponding parts in the P&P library, as things like vision parameters can differ significantly. You probably also want at least "thick" and "thin" variants for capacitors.

 

[LabSpokane]

RX8,

What drove you to in-house P&P?  Is the cost of doing small runs on the outside so high that this is a much more economical solution?  I'm not questioning your decision, just curious. I've not had to walk this path on my own yet, but its coming.

[rx8pilot]

RX8,

What drove you to in-house P&P?  Is the cost of doing small runs on the outside so high that this is a much more economical solution?  I'm not questioning your decision, just curious. I've not had to walk this path on my own yet, but its coming.

Its a fair question. I have been making small batch specialty PCB's that tend to change a lot. LOTs of prototypes, and pilot runs. As the designs have become more mature and sophisticated, they have become more dense and use more fine pitch parts. Hand assembly was getting harder by the second so I looked for a few assembly houses to see what the options were.

The good news is that there are quite a few small assembly shops a short distance away. I met with a few and showed what I have and got quotes. The prices were fair, but the financial penalty was rather severe if I wanted to run 10-20 pcb's on a quick turnaround to do some tests. It did not make much sense financially until I got to around 100pcs and could wait 2 weeks for them. In addition, I would have to kit the project and document all the details which takes a considerable amount of time. What I really wanted, was to have complete control and flexibility to to prototypes with BGA, 0201's, double sided, etc put the PCB on the bench and shake it down. Order the next revision of the PCB and put it back into the same setup on the machine and run 25 more within a few hours and on my schedule.

When faced with the cost of a 'real' machine it was hard to justify. I was expecting to spend well over $50k for a basic machine. I already have a stencil printer and batch oven so I only needed a machine with feeders. The Quads looked like a great place to start because I can pile on a TON of feeders so it can be setup for multiple PCBs at any given moment. They are also supported by PPM and a few others. When I found this machine for only $5k including the feeders, I was able to snatch it up quickly. By the time I finish, I will have put in another $2500 or so for parts plus about 60 hours of labor. Really not bad at all for the ability to assemble PCB's at any level of complexity I will ever see. With a rating of 3600CPH (maybe 3000CPH in real life), it will outrun our sales for the foreseeable future too.

I did the same thing when I built a CNC machine shop nearly a decade ago. It was awesome to have the flexibility and total scheduling control over our parts. Most CNC shops (understandably) can't deal with someone barging in that needs parts or a prototype in a few hours. Having these things in-house was also a FANTASTIC learning experience as a design engineer. I actually programmed, setup, and ran the machines. I made fixtures and tools. I studied the process in detail. When I sat down to design a new product, I knew exactly what colors I could paint with. I knew what was easy and what was hard.

This is a bit of a ramble....hope that it answers the question. I am guessing that the cost of the machine will be covered in about 3 months, the labor I have put into it will be returned in about a month. That is about the fastest ROI I could have dreamed of. The time savings will be split between new designs and sales efforts. So, I see this purchase as a critical pivot point in my business that will allow us to get over the hump and grow up.

[LabSpokane]

That's a totally legitimate answer and it sounds like you've really thought through the "payback."  Thanks!

[mikeselectricstuff]

RX8,

What drove you to in-house P&P?  Is the cost of doing small runs on the outside so high that this is a much more economical solution?  I'm not questioning your decision, just curious. I've not had to walk this path on my own yet, but its coming.

Its a fair question. I have been making small batch specialty PCB's that tend to change a lot. LOTs of prototypes, and pilot runs. As the designs have become more mature and sophisticated, they have become more dense and use more fine pitch parts. Hand assembly was getting harder by the second so I looked for a few assembly houses to see what the options were.

The good news is that there are quite a few small assembly shops a short distance away. I met with a few and showed what I have and got quotes. The prices were fair, but the financial penalty was rather severe if I wanted to run 10-20 pcb's on a quick turnaround to do some tests. It did not make much sense financially until I got to around 100pcs and could wait 2 weeks for them. In addition, I would have to kit the project and document all the details which takes a considerable amount of time. What I really wanted, was to have complete control and flexibility to to prototypes with BGA, 0201's, double sided, etc put the PCB on the bench and shake it down. Order the next revision of the PCB and put it back into the same setup on the machine and run 25 more within a few hours and on my schedule.

When faced with the cost of a 'real' machine it was hard to justify. I was expecting to spend well over $50k for a basic machine. I already have a stencil printer and batch oven so I only needed a machine with feeders. The Quads looked like a great place to start because I can pile on a TON of feeders so it can be setup for multiple PCBs at any given moment. They are also supported by PPM and a few others. When I found this machine for only $5k including the feeders, I was able to snatch it up quickly. By the time I finish, I will have put in another $2500 or so for parts plus about 60 hours of labor. Really not bad at all for the ability to assemble PCB's at any level of complexity I will ever see. With a rating of 3600CPH (maybe 3000CPH in real life), it will outrun our sales for the foreseeable future too.

I did the same thing when I built a CNC machine shop nearly a decade ago. It was awesome to have the flexibility and total scheduling control over our parts. Most CNC shops (understandably) can't deal with someone barging in that needs parts or a prototype in a few hours. Having these things in-house was also a FANTASTIC learning experience as a design engineer. I actually programmed, setup, and ran the machines. I made fixtures and tools. I studied the process in detail. When I sat down to design a new product, I knew exactly what colors I could paint with. I knew what was easy and what was hard.

Totally agree with all the above - the reason I got a P&P was that it was a pain to deal with subcontractors for small jobs, and I often needed fast turnrounds (like design-start to delivery under a week).
In my case it wasn't much about cost as most of my customers could have  taken that, but about the time & hassle of kitting and documentation.
There can also be issues of only finding out late that there is a problem - e.g. in on case a part got missed off the BOM, so didn't get fitted - if I'm assembling a board I designed a few days ago, I'd have spotted it straight away.

Also totally agree about the learning process - seeing the assembly process close-up is valuable input to future designs.
When I do subcontract out larger jobs, my local place now doesn't charge me any setup fee as they know it will all be ready to run.

Another issue I had was that subcontractors would only use their normal full-size metal stencils at GBP170-odd each, whereas I can use smaller frameless metal or plastic ones for small runs.

[Corporate666]

RX8,

What drove you to in-house P&P?  Is the cost of doing small runs on the outside so high that this is a much more economical solution?  I'm not questioning your decision, just curious. I've not had to walk this path on my own yet, but its coming.

I know you didn't ask me, but if I may just horn in with a real world example.

I used a Chinese based PCB assembly service in years past.  The cost of assembly is always hard to quote without looking at the board because there are so many variable, but anyway, this particular PCB house has a standard base rate of $0.0125 per pad (regardless of single or double sided, or large or small parts).  So 1.25 cents per SMD pad.

I have one product that just has a bunch of PLCC-4 LED's on it, some passives, a couple of chips and associated regulation.  30 individual parts in total.  There are 100 pads total on the board (SOIC-8 is 8 pads, for example).  So that board would cost me $1.25 to be assembled.  There is also a setup fee of something like $150 or thereabouts.

I used perhaps 5,000 of these boards per year.  Based on the setup fee and assembly price, it makes sense to order around 1,000 of these assembled boards at a time.  That entire production run will use a half reel of some parts, 2 reels of other parts, 10 reels of LED's. 

With all due respect to the Chinese, I don't trust them sourcing parts for me at all.  Even when I have specified (for example) a National Semi branded 7805 regulator, I've had staggeringly high failure rates and do not believe the chips were genuine for a moment.  And passives are so cheap that it is absolutely foolish to try to save $4 on a production run using Chinese sourced parts.  So long story short, I would buy reels and ship them off to the fab in China.  That means the $2,000 I spent on a reel of 2,000pcs SOIC-8 microcontrollers will only get half used... and it will be sitting in a factory in China for 2-3 months until my next production run. 

Net/net, my run of 1,000 PCB's will cost me $4k or $5k when you add in the cost of all the parts, setup fees, assembly fees, shipping both ways, etc.

I have three Quad 4C's that I paid around $6-8k each for.  In the course of a single year, the cost of the machine is more than paid for just by what I would have spent on those 5,000PCB's - which would have been over $6k, before setup fees, shipping costs etc.

The additional benefit I get is that I can buy my reel of $2,000 SOIC-8 microcontrollers and leave it on the machine - and if I need to do an emergency run of 50 PCB's, it's easy.  If I need to grab 5 of them to repair PCB's that were bad, it's easy.  If I have another product using the same parts, I don't need to inefficiently have thousands of them sitting on a shelf in China while I am hand assembling other boards with more of the same parts I had to buy separately. 

And it's not just high value parts like MCU's... even "cheap" parts like power inductors at $0.25/ea get expensive when they come in reels of 3,000pcs and there are several equally expensive parts on a board.   

So in a nutshell...

-Saves money
-More flexible to your needs
-Allows you to buy parts "just in time" vs warehousing dozens of reels of parts

[LabSpokane]

Corp666,

No problem. Thats great reasoning. Thanks for sharing.

[Towger]

How is the machine going?

[rx8pilot]

The machine is going great, however, I am unfortunately struggling to make the time to get to the next stage - programming.

There are 'emergency' orders that I have to work on, the kids are on spring break, and I am also laying out new PCB's. Next week I have a trade show that will consume even more time. Anodizing went bad on our last batch of enclosures so I have to tend to that.

Frustrating having it sit there while I build things by hand. I need a few straight days to get it rolling I think. Learning the software and doing tests. LOTS of parts to be loaded.

[rx8pilot]

I have some time coming up that I am planning to use to get the PnP online. The next step is setup and programming. This may be the hardest part. There is no forum that discusses the PPM version of a Quad IVC machine, so I have to rely on PPM or hacking to figure it out. Hacking is slow and risky while PPM is faster but very expensive.

I have gone through the print manual and a you tube webinar to get the basics. The problem is that it simply goes from menu to menu without any real presentation of workflow or concepts. Its like giving an oscilloscope manual to someone that does not know the fundamentals of test and measurement. The manual is actually called a "programmers reference". All I rally want right know is an example project to run through so that I can understand the steps and the process.

So....to hack or not to hack. I don't have much time or money at the moment so I need to choose wisely.

[Corporate666]

I have some time coming up that I am planning to use to get the PnP online. The next step is setup and programming. This may be the hardest part. There is no forum that discusses the PPM version of a Quad IVC machine, so I have to rely on PPM or hacking to figure it out. Hacking is slow and risky while PPM is faster but very expensive.

I have gone through the print manual and a you tube webinar to get the basics. The problem is that it simply goes from menu to menu without any real presentation of workflow or concepts. Its like giving an oscilloscope manual to someone that does not know the fundamentals of test and measurement. The manual is actually called a "programmers reference". All I rally want right know is an example project to run through so that I can understand the steps and the process.

So....to hack or not to hack. I don't have much time or money at the moment so I need to choose wisely.

There aren't any placement programs left on the machine???

If not, surely PPM would send you one, no?  It would be ridiculous if they would refuse to send you a sample program "so you can test out your machine".

[rx8pilot]

I inquired about 1-on-1 training options to get a sense of the work flow. Its a four hour minimum which is $800. Was really hoping for a more flexible answer.

I cannot imagine why it would take anywhere near 4hrs to get a basic setup and program going.

[mikeselectricstuff]

I doubt it is really that hard - If you can't sit down with a real job and a days' time and have it mostly working, PPM have done  a poor job.
It's not rocket science - pick up a part, put it at <x,y,theta> repeat til done.
There will be quirks, shortcuts etc. that you pick up over time, but it shouldn't be hard to get the basics up & running in a few hours.

[Corporate666]

I inquired about 1-on-1 training options to get a sense of the work flow. Its a four hour minimum which is $800. Was really hoping for a more flexible answer.

I cannot imagine why it would take anywhere near 4hrs to get a basic setup and program going.

For training, yes... but I'm sure they would be able to send you a sample placement program that you could look through.  I'm not sure how things are done in the Windows GUI, but on my machine, you define pickups, placements, and then a program.

Pickups are just like an excel spreadsheet, with column titles of "Number, X, Y, Z, Length, Width, Height, Description, Alternate".  The number is arbitrary and is just what is used in the placement program, so for example if I just put a feeder anywhere on the machine, I then move to it with the hand control and place the component directly in the center of the crosshairs.  I then press the camera offset button on the hand control which moves the nozzle to where the crosshairs were on the screen.  Then I jog down in Z and press the Vacuum toggle button and keep jogging down in Z a few thou at a time until I don't see the part jump up when I turn on vacuum.  Then I just basically hit enter and the X/Y/Z coordinates of that pickup are stored.  I use a set of calipers to measure the length/width/height of the part.  They are entered in thousandths... so 0020 is 0.020".  I enter whatever number I want to call that pickup, let's call it 1.  And I enter description, like "10k 0805 Resistor".

Pickups are independent of the placement program.

Then I create the program to place parts.  I start by defining my placement locations.  Almost the same as setting pickups.  I take a PCB and put some thin double sided tape on it, then load it into the machine as if I was going to place it.  I can hit "pickup" on the hand control, it then asks me a pickup #... we programmed "1" before, so I hit 1 and press enter.  The machine will go pick up an 0805 resistor. 

Then I just got the head over to where I want to place the part and center it using the crosshairs on the camera.  Then I press the button to switch from camera to nozzle which just moves the machine the X and Y offset amount so the nozzle is where the crosshairs were.  Then I jog down in Z until the part touches... then just go down a little more - like 0.010" (until I just start to see the spring on the Z axis start to compress).  Then I hit enter and that is my placement location.  Then I repeat the placement teaching process for each location I want to put a 10k 0805 Resistor.  In reality, I don't really need to reprogram the Z setting, and if I keep track of the X and Y coordinates of that 1st part, I can set a "MARK" in Eagle at that location, and then just look at the relative coordinates of each other component and rather than teaching those locations, I can just enter them by hand as new lines in the placement screen.  Keep in mind the X axis is front-to-back on the Quad (at least on mine - X positive means head closer to you) and Y positive is head more to the left.  Home is back right on the machine, so all your X and Y coordinates are positive from that home position.

You also define nozzle locations the same way you define pickups - a number, an X/Y/Z coordinate - and that's it.

For the placement program, it's ultra simple... you have already defined pickup locations, placement locations and nozzle locations and assigned them easy to remember numbers.

Your program then becomes

NOZZLE 1
PICK 1
PLACE 1
PICK 1
PLACE 2
PICK 1
PLACE 3
...
...
END

That program places a 10k 0805 resistor at location 1, then 2, then 3 and so on.


I realize my machines are DOS and yours is Windows, but the manner of operation is pretty much the same.  The Windows version has some part management mumbo jumbo included, but you should be able to work through it.  There isn't much to lose in experimenting... the Z rod bellows just unscrews, and without that, even at maximum down in Z, the head can't come into contact with the PCB or with the feeders.... so if you are worried about crashing the machine, just pull the Z rod and try to write a dummy program.

I'd also press PPM to sent you a test program, or look through the drive to see if there are any on there... I'd be very surprised if the prior owner took the time to wipe out all of his placement programs.

[mikeselectricstuff]

Then I just got the head over to where I want to place the part and center it using the crosshairs on the camera

WTF? seriously? Manually teaching every part position...
Surely your PCB software can output a list of all part positions and rotations, to import into the P&P machine....?

[Corporate666]

Then I just got the head over to where I want to place the part and center it using the crosshairs on the camera

WTF? seriously? Manually teaching every part position...
Surely your PCB software can output a list of all part positions and rotations, to import into the P&P machine....?

Of course it can.... but I manually program the first location and then just key in the X/Y deltas off that part.  The reason is that I usually let the PCB house decide on the panelization and they add the rails and fiducials, so I don't have precise X/Y coordinates from fiducial to board.  Only takes about 30 seconds to teach the first location - I don't teach them all

[rx8pilot]

It is true that it is not that hard - in theory. I don't want to bash PPM's efforts but I will be honest with what I have seen so far. Some of what I say should be taken with a grain of salt because this is a new process for me. I make no claims of knowing what I am doing when it comes to PnP operations (yet).

The concepts are rather easy to grasp for sure. This is especially true since I have spent many thousands of hours programming and operating 5 axis CNC machines and designing and programming various motion control systems. Tell it where the part is, tell is where is goes, repeat as needed until the PCB is populated. In general, this machine and it's software are totally capable of that task. The rub is that there is no flow or thought in how it is laid out that I can tell. Normal windows programming conventions are non existant - it makes Eagle CAD look like a graceful butterfly. Keep in mind this is 10+ years old so this should NOT be compared to whatever the current version is today which was designed for Windows 7. Since the mid-80's I have dealt with all types of weird and strange pieces of software. The low-volume specialty industrial stuff has been the most bizarre. Using these pieces of software has given me some skill in getting to know the programmers intentions and getting what is needed from otherwise terrible software. Not everyone can be Apple I guess.

Much of the programming information appears in multiple places but can only be manipulated from one place. It is not obvious where to go to add a feeder when looking at a list of feeders on the screen for example. None of it is unworkable, it is just messy and not very considering of a typical operator. I am struggling to find a better way to say 'NOT INTUITIVE'. I have prepared some CAD data for testing which is simple. I did a ULP for Eagle to format the data exactly the way it is needed for the PPM software. Importing the data is easy enough - desginator, package, X,Y,Theta, etc. The software appears to be operating from a database and not individual files (which seems risky to me). It has it's own organization scheme internally to keep track of user generated data. Now, how do you assign parts from the CAD data to a feeder? How do you tell it where the feeders are? How do you get the PCB oriented in the machine? The answers are all there - just in such a strange way. Something that looks like a button is actually a menu....nuts.

With that said, I turn to the manual. I can't figure it out by looking at it and pushing buttons so the next best thing is reading the instructions, no big deal. This is not at all a sign of bad software design, but more an amateur trying to learn something new. The problem with the manual is that it is essentially a dictionary and not a textbook. The difference is that it rather simply tells me what an individual button may do or what data is expected in a particular box. It does not take you through the process at all or present concepts. It would be nice to have a flow chart of sorts that explains the course steps to be taken from one end of the process to the other. The process in concept is so simple, it is hard to understand why it feels so cumbersome and complicated. I took some time this morning to go and fiddle with it for a bit, and was able to get a board defined on the table, added some placements, added some feeders and nozzles. Convoluted and strange - but the machine seemed happy with the data so far. I can deal with convoluted if the PCB's are getting done. I did not have a functional program, just some X/Y locations.

It all stopped when I was setting up the nozzle changer. I told it to pickup a nozzle but it did not pull it out of the pocket. I am not sure if the nozzle changer did not release it or the 'Z-Rod' that picks it up has bad o-rings. Either way, the nozzle pickup failed and the machine stopped. An error window pops up to tell me what happened. Not shocking at all that the pickup did not work - could have been a number of things. The problem is that the Z rod stopped in the down position, meaning if the X or Y axis moved - it would be damaged. I turned off the machine and manually moved the Z to a safe position, turned it back on and re-homed everything. I tried to do a nozzle pickup again and the same result - it is right on the X/Y target but the pickup fails. This time I was looking for a way to retract the Z to a safe position without turning the machine off. There is a 'HOME ALL AXIS' command that I have used a number of times to re-home the machine as I have been testing the basic movement, joystick functions, etc. It behaves just like a CNC machine. First it retracts the Z slowly and finds its home. Then is jogs the X and Y axis to find home. The problem I found here is that it only behaves that way when there is not already homed. When I pressed the 'HOME ALL AXIS' button, it launched the X/Y home at full speed with the Z all the way down - damaging the Z pickup rod.

That is a flaw. Period. There is no reason the X/Y should ever be in motion while the Z is down. The command should also not have a dangerously different function depending on the status of the machine. Anyway, the Z-Rod is now bent. Nothing else is damaged that I can tell. All next week I have to do a trade show, so waiting on the bent part to arrive is no big deal.

So far, the machine will most likely solve my challenges for PCB assembly which is the good news. There will likely be continuous frustrations with silly software design along the way. I am happy that I have not yet been stopped dead in my tracks - PPM can always bail me out as long as my credit card can stand it. Having support for the machine was a big reason I thought it would be a good idea but as Corporate666 explained, PPM may not be a good business match for someone like me trying to get a tiny PnP machine up and running on a shoestring budget - sub $10k. They are nice and knowledgeable people, but the service and support options and prices outclass the decades old machine. It feels like I am being punished for buying a used machine to save money and they want to get me to pay for a full price machine somehow. That may be a bit over the top, but $1800 for a joystick in a hand drilled off the shelf project box? Minimum of $800 for a one-on-one phone training even if I only need help for an hour? $1000 for a vibratory feeder top plate? I did pay $1500 for the software licence because they say the previous owner never paid for it to be licensed. That is fair enough, not a problem at all. When I asked if there were any maintenance updates, they said that was going to cost extra. That was the answer without even knowing what version of software I was on. I would think that the freshly paid licence should at least be the latest version available for my particular system right? Anyway, I am thankful that they are in business but they make me want a fresh and new machine already. If I am going to have the brand new machine costs, I would like a new machine. If I have a 20+ year old machine, it's because it's all I can afford and I am clearly looking for a low cost solution. That is not a knock at all toward PPM, it's simply a reminder that entry-level PnP does not really exist without some pains along the way. There are most likely many machines out there that are installed and working and it is easier to just pay to maintain an OLD machine than get a new one. I cannot imagine that will last too much longer as these things are rather old the maintenance costs will capture the attention of the bean counters.

If I had PPM do all the service needed to get this machine going, I would have spent $10k+ by now in addition to the $5k I paid for the machine. So far, I have spent about $2k and did the other $8k myself in about 40hrs or so - nights and weekends. I don't want to be a machine mechanic, but for $200/hr it's not horrible. I could almost buy two spare machines for that much money.

I have to remind myself that I only have to make a small qty of PCB's for this to be a good decision. If I make a medium qty of PCB's, it will be a great decision.

[Corporate666]

You are running into the same experiences I had with PPM.  Nice enough people, but they are batshit crazy in their prices.  They are trying to command premium prices in a market that is dwindling fast.  They don't have any new offerings because although they bought the rights to Quad/Tyco's manufacturing and tooling, they are not capable of making PnP machines... so they are clinging to a shrinking market of legacy customers who need support.  Their solution to that was the Windows upgrade.... but rather than hire in-house talent or go with a hot-shot programmer, it seems they went with someone who was a bit of a hack and got the job done, but not well.  That, combined with the prices they are asking for this machine, makes it a tough sell.  They sell it as a brand new machine, but it's not.  It's a custom add-in card on top of a 20 year old machine.  The machine still uses the old motors, old power supply, old vision system, old servo drivers and old mechanicals... it's like buying an older car with a fresh paint job.  It ain't the same as buying new.

I wish they had decided to try to foster a new market by open sourcing the software and they could let the community improve it - and then a Quad 4C would be *the* PnP for the small shop to buy - with the best software and support.  That would raise values on the machines and make PPM the place to go for a machine or for parts.  But they went a different route - to milk as much as possible from the legacy users - virtually guaranteeing those customers look elsewhere when their needs progress.

But I digress...

You're correct that allowing movement in X or Y while the Z is down is absolutely asinine.  (not to say I told ya so but that's why I suggested removing the Z rod)

Anyway, if you haven't already ordered the Z-rod, go with these guys

http://www.nozzlesupply.com/quadnozzle.htm

I bought a couple of Z-rods from them - they were cheaper than the PPM ones, and they were brand new and an improved design.  My OEM rods would clog up with solder paste over time and get gummy... the one from these guys has literally hundreds of thousands of parts on it and it has never bound up or gotten stuck even once.

Also, I was looking through my prior purchases and came across these guys..

http://www.nozzles4smt.com/Count-On-Tools-Expands-Its-Quad-Feeder-Peel-Roller-Series_b_60.html

They sell replacement urethane feeder rollers for Quad feeders.  They are quite a bit less expensive than OEM/PPM.


As for your nozzle failure... does the Z rod have the o-rings on it?  There should be 2.  Did the nozzle changer slide-lock pull back to release the nozzle?  It sounds like it might not have.  Check the connection (easily accessed by pulling down the access panel just under the table). 

[mikeselectricstuff]

This thread is starting to sound very familiar.....
There is a similar situation with the old Versatronics machines, mostly sold in the UK. Despite being about 20 years since the company went titsup, it is probably still the smallest and best value P&P available for the low-end user. Used machines sell nowadays for £1000-4000 with feeders, it needs no air, and can be transported by one person in a small van.
I bought mine a few years ago from a guy who used to work for the company, and had cornered the market in supporting the existing user base. This guy charged for everything, and wouldn't hesitate to screw as much money as he could. As soon as he had the money it was almost impossible to get hold of him, even if you wanted to buy parts etc. 
If you'd just bought a machine he might grudgingly tell you a few tips and try to make you think he was doing you a huge favour.
So I set up a forum for machine users. As it's a very small  user base, this takes negligible time to run.
This really pissed him off. He registered and participated briefly but then resorted to registering new  usernames and trolling.
He's not been heard from for the last couple of years, and the users of the machine are now pretty much self-sufficient - beginners get a helping hand dealing with quirks, software utilities, tips  and sources for parts get shared freely, a guy with a  lathe makes new nozzles. Someone else has re-created some of the PCBs used in the system and sells them at reasonable prices. Someone else is currently reverse-engineering the software to allow it to run on newer PC hardware.

The situation with PPM is slightly different in that it is their software, but in terms of support, other users will be at least as good in helping deal with issues that PPM might want to sell training to deal with.
As the market for the Quad moves downward towards  hobbyists and low-end users, community support will be increasingly important.
As it is, everyone will go through the same learning process in getting the first job set up - if one used documented this it would be hugely useful to others.
So maybe an answer is to set up an independent user forum.

[Kjelt]

Concerning PPM:
The prices for the hardware are bullocks but a one on one training from a company, I have seen worse prices.
A normal course with 20 participants can go up to $125 an hour per participant. So not sure if that is so bad.

[Corporate666]

The situation with PPM is slightly different in that it is their software, but in terms of support, other users will be at least as good in helping deal with issues that PPM might want to sell training to deal with.
As the market for the Quad moves downward towards  hobbyists and low-end users, community support will be increasingly important.
As it is, everyone will go through the same learning process in getting the first job set up - if one used documented this it would be hugely useful to others.
So maybe an answer is to set up an independent user forum.

There is a pretty good user base over at SMTNet but you're right, an independent forum specifically for the 4C would be ideal.

Regarding rx8pilot's machine... the more I hear about the software, the more I feel that sticking with the DOS machine is the way to go.  There is a much larger installed base of DOS machine users, and considering the Windows version was created solely by PPM, it means you have a single place to go if the shit hits the fan as opposed to being able to get help and support from many users.  Not to mention that the additional features aren't in the basic capability of the machine, just software add-ons, like part management.  But the 4C was designed from the get-go to be a flexible and adaptable machine, so adding stuff like BOM management goes against the nature of the machine's basic design, IMO.

I believe PPM has already seen the writing on the wall as they have occasionally contacted me about the Windows upgrade and soliciting the sale with offers of reduced pricing, etc.


Regarding the "charge up the arse" mentality...

A few years back, before I bought any machines, I talked to PPM who wanted $24k for a Windows machine, $19k for a "refurbished" DOS machine, or ~$13k for an old machine that has been tested and works, but no refurbishment.  I later bought 2 DOS machines from a medical company in Arizona.  The machines were missing a few parts, but nothing too bad (missing the Z rod's for example).  I later bought a third machine which was the PPM $19k model (refurbished and upgraded with LCD monitors, etc).  I did not buy from PPM, but from a local electronics company moving to China production.  They had paid in the $30k range for the machine - I got it for $6k with all the feeders, lots of accessories and they spent a couple of hours showing me how to use it before I loaded it up.  I later contacted PPM about doing the Windows upgrade which costs $7500.  I inquired about trading in both my older DOS machines in exchange for the upgrade and was told the older machines aren't really worth anything - they could perhaps give me $1k off the price of the Windows upgrade.

So in summary... the machine they sell for $25k-30k has a fair market value of around $6k-8k.  That is just simply what they sell for in good running condition with feeders.  The Windows upgrade for that machine is $7500, but they require you to send them back all the old parts you swap out with new parts - like the control panels, control pendants, the main control board, etc.  They will resell those parts for a sum total far in excess of $7500.  And the machines they will charge $13k for, they will not offer more than a few hundred bucks in trade-in and won't buy at all straight out for cash.

Draw your own conclusions 

[rx8pilot]

Concerning PPM:
The prices for the hardware are bullocks but a one on one training from a company, I have seen worse prices.
A normal course with 20 participants can go up to $125 an hour per participant. So not sure if that is so bad.

I am not upset with the hourly rate, I just did not want to commit to 4hrs. If it actually took 4hrs, so be it and I pay the $800. If I get what I needed in 1 hour, I would pay $200. Was only hoping for flexibility. If the manual was written well, there would be less of a need for $200/hr training.

Regarding the "charge up the arse" mentality...
..........
Draw your own conclusions 

I will never fault anyone or any company for making money. All I am saying is that it feels like the ship has sailed on these decades old Quad machines for high-end users. I could actually build a 'clone' of this machine from the ground up with new and modern electronics and be seriously profitable at $28k per machine. In reality, I have no aspirations to be a PnP manufacturer, but just to make a point. I have put my eyes on every detail of the machine so far and have every skill and machine needed to construct a fresh version of it with 'entry-level' cost as a target. It would be brand new.

The situation with PPM is slightly different in that it is their software, but in terms of support, other users will be at least as good in helping deal with issues that PPM might want to sell training to deal with.
As the market for the Quad moves downward towards  hobbyists and low-end users, community support will be increasingly important.
As it is, everyone will go through the same learning process in getting the first job set up - if one used documented this it would be hugely useful to others.
So maybe an answer is to set up an independent user forum.

Could be a good idea and it would not be too hard. Building some community around the system could extend the life of it and make it much more friendly to low-budget start-up types of businesses.

[rx8pilot]

PPM just sent over an updated PDF manual. I have to say.....that is awesome. The one I had was missing many of the images, but this new one is complete. Pretty happy about that.

Going through it now.

Just registered:
quaduser.com
To make a Quad specific discussion forum. Hopefully it will make these machines more accessible to new low-budget users like me.

[rx8pilot]

Headed to the shop to CNC some fixtures for the PnP. The converyor system that shuttles the PCB's in and out of the machine can only do rectangular shapes. I wanted the flexibility to use the conveyor system to hold odd shaped pre-routed PCBs that are double sided. I did not find any commercial product that fit the bill so I designed one that fit exactly what I wanted. The PCB can be suspended in the middle or justified in a corner if it has two edges at a right angle.

This will allow the machine to accommodate regular panelized PCB's as well as the weird 1 off prototypes without having to change anything. Right now, I have quite a few pre-cut boards that will not fit the conveyor.



I am also making a JEDEC tray that is a cut-tape holder. It is adjustable in 4mm increments so it can do any type of tape at any length. I had found one similar but it was about $500 and I wanted at least 6 of them to be able to pre-load for various projects. Also, the commercial one that I found could not do tall components. There are other cut tape feeders but they are fixed widths - like 8 lanes of 8mm OR 5 lanes of 12mm. I wanted any combination of tapes to be quickly loaded into a tray holder. Being an ME and machinist is helpful here.

The rough un-finished model.....trying to keep the features simple so the CNC part is easy.



I was at a trade show all of last week so PnP is low on the priority list. I did, however, go through the new PDF manual from PPM. It is WAY better than the one with missing pictures. I finally have a handle on the programming. Did some tests with setup and programming so I should be in pretty good shape now. This manual has a "Getting Started" section right up front which is hugely helpful.

I will take some photos at the machine shop. Seeing metal flying is always entertaining.

[JuKu]

The feeders look very nice, but maybe too simple. There are some issues to watch for:

- I assume there will be some kind of top cover? The small parts are very light. Sometimes I feel that an angry look is all that is needed to make parts jump out of their pockets. In other words, unless something is holding the tape down or you have a clever cover tape peeler so that only one part at the time is exposed, parts will jump out from nearby pockets.

- Can you adjust the pickup height and/or downforce? Paper tape is not strong from bottom side, it is easy to press parts through.

I'd like to see these feeders work.

[rx8pilot]

You are correct and it may be a challenge with small parts jumping. There is an acrylic cover that will restrain the tape as well as a pin to help with consistent placement index. The paper tapes will all go into regular feeders. This one is only for plastic at the moment which is where the expensive parts are.

I have the option of putting an adjustable bottom support for the tape, but it adds complexity so I will test without that at first. The machine has adjustable pickup height and speed which may help. Big or heavy parts are not a problem, but a QFN-28 is pretty small. fingers crossed that a gentle pickup should be ok.

Could be an option for your machine if it works. I will be happy to send the data to you.

[Stogiesaurus]

Hi All,

I ran across this forum and thread looking for info on cleaning the vacuum lines of our Quad ??C (not sure which model it is) and decided to sign up.  It's been loosing vacuum for years and I feel it is getting critical.  I was told that it's due to the lines getting dirty and that they need to be flushed out with alcohol.  I hope to see if that's true in the next few days.  I've been using it for close to 10 years so if anyone has any specific question please feel free to contact me.

If the JEDEC tray works out I may be interested in acquiring one from you.  I've thought about doing something similar for years.  Mine would be more like taping the tape to the table sort of like the vib base mod in the photo.   

Thanks,
Larry

[rx8pilot]

Larry,

Glad you joined the discussion. I am not sure what the condition of the vac system is on my unit since I have yet to setup a complete program. The last few weeks,I have been buried with other tasks - design, PCB layout, warranty service on our old products, trades show, etc....

Are you feeding the machine air or vacuum? What do the gauges say? Have you pulled the Z-rod to check for clog?

Hoping to get the JEDEC tape tray out of the CNC machine this week along with the PCB holder. Working VERY long days right now which is a bad time to operate a 30HP CNC machine. Gotta be fresh for that type of work or it will tear my limbs off.

[Corporate666]

Hi All,

I ran across this forum and thread looking for info on cleaning the vacuum lines of our Quad ??C (not sure which model it is) and decided to sign up.  It's been loosing vacuum for years and I feel it is getting critical.  I was told that it's due to the lines getting dirty and that they need to be flushed out with alcohol.  I hope to see if that's true in the next few days.  I've been using it for close to 10 years so if anyone has any specific question please feel free to contact me.

If the JEDEC tray works out I may be interested in acquiring one from you.  I've thought about doing something similar for years.  Mine would be more like taping the tape to the table sort of like the vib base mod in the photo.   

Thanks,
Larry

Hi Larry,

As RX8 said, are you supplying air or vacuum to the machine?  If air, then over time the residual oil from your compressor will accumulate in your machine.  The places to check and clean would be the vacuum venturi.  If you open the left side of the machine (left as you are standing in front of it looking at it), you will see the plumbing inside where the air enters.  There are a couple of very simple couplings/switches.  You can very easily disconnect the vacuum venturi and soak it in alcohol overnight.  Ideally, disassemble it as much as possible and remove any seals - and clean it that way, it will work better.

As said above, check your Z-rod and your nozzles.  Both will accumulate solder paste that will eventually dry and reduce vacuum.  Nozzles are easily cleaned by dropping them in acetone.  Z-rod can be done the same way, but remove all the seals first (or even better, just use a wash or spray bottle with alcohol or acetone). 

I would think it's unlikely that it's the lines themselves, but if you pull the panel and see amber colored slime in your lines, that's compressor oil.  You could pretty easily replace the lines - measure your tubing and get replacements at www.mcmaster.com

[Stogiesaurus]

I quoted RX8's reply yesterday.  I don't see it now so I'm not sure what happened to it.

I clean the Z-rod from time to time.  I realized I needed to do it again after RX8 mentioned it.  We run so sporadically that we don't have a specific maintenance schedule.  It gets done when things start messing up.   

I cleaned up the venturi device yesterday.  It was full of oil but my guess is that is has been for a long time so I'm not sure how much this will help.  If I remember correctly there is an inline air filter somewhere in the head.  I'm going to try to find and replace it.

[rx8pilot]

The venturi is the most likely spot to lose vacuum performance I would think. The compressed air never goes through the machine - only the venturi. Hope that helps. I was considering a getting a vacuum pump instead of running compressed air. seems more efficient since the compressed air is always bleeding out the venturi.

[Kjelt]

If you do buy a good vacuum pump install a good vacuum filter , they are rather cheap and can be exchanged on a yearly basis, helps the lifetime of the pump

[rx8pilot]

Any suggestions on a vacuum pump? I  know it will need 25in/Hg but not sure what volume. Does it need a reservoir?

[Corporate666]

Any suggestions on a vacuum pump? I  know it will need 25in/Hg but not sure what volume. Does it need a reservoir?

My two chipshooters both used vacuum pumps made by Gast - which are about as good as it gets.  I still have a couple of spares, but they are 3 phase.

Neither had a reservoir - the amount of vacuum used is very small and the line volume is enough to prevent drops in vacuum before the pump can kick in.

[Kjelt]

AFAIK you have two sorts of vacuum pumps. The very deep vacuum that go to -0.99 bar at the end but they have not much capacity so takes a long time.
And you have the high capacity types that will not go below about -0.9 bar.
First see which one you need for your application, for cnc machines and i think for this application too it is the latter.
Look for maintenance free vacuum pumps or you have to change the oil each year ( moist in the oil)

[LabSpokane]

Out of curiosity, just how much air/vacuum do PnPs need?  I would think that a small air ejector would be more than sufficient.  I'm very surprised to hear that a three-phase vacuum pump would be required. 

[rx8pilot]

I don't know the exact figure, but it is small. The air compressor/venturi method seems wasteful since it is always bleeding air to maintain vacuum. It also needs the air to be appropriately filtered to keep the machine from clogging. Water removal, oil separation, and particulate filters are all separate pieces.

At my CNC shop we had a high-end air system the was super filtered, but I only have a small 1HP/3 gallon compressor where the PnP is that I used for manual solder paste dispensing and cleaning stencils. With the PnP machine just sitting, that compressor cycles every few minutes. Guessing that it would run continuous if the PnP was going.

Hoping that a vacuum system will only run when the machine is cycling and I would never have to worry about the machine getting clogged on the supply side. I will get solder paste every once in a while at the head, but that would happen either way. As Kjelt mentions, a maintenance free high-volume single phase pump should be fine. Does not take much to hold an SMD component, right?

[LabSpokane]

Gotcha.  I just calculated the CFM on a small ejector and it's 0.7, so that would be really annoying with a small compressor.  Even a 5HP compressor would be cycling on and off all the time.  I was so spoiled (and lazy) in a past life with room full of big axial screw compressors.  We just didn't think twice about using an air ejector for our end effectors. 

Thanks again for updating the thread with your experiences.  It's a good learning experience. 

[Corporate666]

Out of curiosity, just how much air/vacuum do PnPs need?  I would think that a small air ejector would be more than sufficient.  I'm very surprised to hear that a three-phase vacuum pump would be required.

Well, you haven't seen the machine )))

Rotary head chipshooter with 24 heads that places over 18,000 parts per hour.  The vacuum pump is sort of a shotgun approach to the problem... what I mean is, rather than having perfectly sized nozzles, well-maintained rubber seals and worrying about the amount of vacuum needed for various sizes and weights of components, they just use a really strong vacuum which will overcome sealing issues and ensure the parts stay in place as they whip around on their way to the PCB.

[rx8pilot]

I would be happy placing 18,000 parts per month.

[LabSpokane]

Ahh...now I have understanding! 

Another ignert question:  does the machine just release the part by releasing vacuum or does it put some positive pressure in the line to speed the process? 

[rx8pilot]

That is a good question and I am not the expert. From what I have examined, the valve is rather close to the nozzle so a simple close the valve maneuver is enough to release on the Quad IVC. The PPM software (maybe DOS as well) allows the programmer to specify how long to wait for pickup and release. In practice I don't really know how necessary that will be. 

[Stogiesaurus]

If you only need compressed air for the PnP a vacuum pump might be the way to go.  Have you done the power consumption calculations?  We have about a 40 to 50 gallon tank on our compressor and it only runs every hour or two.  Without a tank I would think a vacuum pump would need to run continuously.

ETA: Our Quad doesn't have a puff option to blow off the part.  It would need a source of air for that.  I think there are some newer models that do but maybe not in the C series.  It is important that you push the part down into the paste.  Otherwise the parts wind up all over the board. 

[SeanB]

A few notes on vacuum pumps. You can get a Gast oilless pump, and it will work, but you must use the GAST filters before the pump and after it, as they really do not like ingesting anything particulate. I speak here from experience, where one picked up some thread slivers and ate itself. Plus they really want a prefilter for the filter, as those are expensive as well. I used a Donaldson truck air filter and the secondary filter, so that the Gast filter stayed absolutely clean ( it went in and came out a year later looking like the other one in the sealed pack) and happy, while the massive pleated filter was busy picking up all the dust and dirt, giving 5 micron air out to the pump.

Those Gast pumps really do not like moisyure either, or not being run regularly, as the cast iron bores rust, and then you grind the graphite vanes to pieces. Those then make great cup holders.

If you want a cheap vacuum pump buy one of the Chinese ones with a separate motor, preferably with a 3 phase motor, or be prepared to be replacing the single phase 1.5 horsepower one on a regular basis as it cooks itself ( single phase induction motors run so hot and are so cheaply made that they fail after a while, but the 3 phase one lasts pretty well if it makes a week) and also change the oil on a regular basis, along with routing the exhaust air outside as you do not want to breathe that oil mist. Those are pretty robust, and with a oil coalescing filter on the output so they recover most of the oil and reuse it. Inlet you want a good vacuum filter, so that you do not get debris into it.

For light duty you can use a refrigeration vacuum pump, with the pump and motor all in one unit. Low flow, very good vacuum and good for light use, and the same about oil and filtering.

Emergency vacuum pump I have used ( waiting for the Gast to come from the USA) a canister vacuum cleaner, which actually gave good enough results. Noisy, but got down to 0.1bar abs and had enough flow to cope. Ran for around 2 weeks 5 hours a day, and still is working. I bought a spare motor just in case, but never used it there. It went in another canister vac.

Regarding valves, you absolutely want to use piloted valves with vacuum, and preferably the biggest ports you can get, for a pick and place use a 1/2in port valve, as then it will have a fast pick up and fast release, and have it as close as possible to the pick up point. If it is a small suction cup you also want to have a small filter in the line to the valve, so use a very small fuel filter ( 6mm inlet and outlet, with as small a volume as possible) so the valve does not pick up debris and start to stick. That will happen fast with direct valves, the piloted ones will have enough force to handle small amounts of dust without too much issue for a while.

[Corporate666]

Ahh...now I have understanding! 

Another ignert question:  does the machine just release the part by releasing vacuum or does it put some positive pressure in the line to speed the process?

On the chipshooter, it uses positive pressure to blow the part off and into the solder paste, but on the Quad IV-C it doesn't, it simply pushes the part into the solder paste and relies on the stickiness of the paste to hold it.

The chip shooter machines are really something to behold... check this out (the porn starts at 00:45)



They move so fast that the mass of the parts, becomes an issue since the board is accelerating so quickly around in the X and Y axes, and the rotational head is moving so fast that a fair bit of vacuum is required to hold the parts.  This video isn't my machine, but it looks and functions exactly the same... on mine, it would adjust it's speed based on the mass of the part it picked up (so it didn't sling it across the shop), which is one reason why the speed at which the heads rotate changes - that and to give the PCB time to get into position.

[rx8pilot]

A few notes on vacuum pumps.......

Thank you Sean. I think the Gast pumps are an easy choice and will be an upgrade from a compressor/venturi. They are plentiful here in California new/used. When I built my CNC shop, I started with crap air system and ended up paying for it with machine damage and lost time. I ended up with a very high-end variable speed rotary screw from Atlas Copco. It was actually connected to the factory via a cellular data link to report maintenance data. At 175psi and 70CFM it was about as loud as a refrigerator. Peace of mind.


Once again I am headed to the CNC shop to attempt to make the fixtures for holding PCB's and cut tapes. A few updates to the JEDEC tray cut tape holder. I added a spring loaded retractable registration pin to help load the tapes accurately and prevent them from moving on accident. I also split the rails in half so that I can have the option of having a 12.7" long lane OR a 6.2" long lane. In the rendered image you can see a full length 32mm along with half length 44mm, 24mm, 8mm and 12mm. It had occured to me that with a QFN microcontroller, a full length of them is too many. This is pretty flexible. Also working out a bottom support for the wider tapes. FYI - not intended for paper tapes.



Hoping I can actually get something done today. So many other projects keep jumping in front.

[rx8pilot]

After a major delay, I got back into this machine. I have been pulled in every direction while this thing sat around. As my new PCB designs are proving to be going toward production, I needed to get this thing going.

On August 3, it placed it's first part. Super excited bout that. One 0805 resistor on an old PCB I found in a scrap box was somehow totally awesome. I have spent about 100 hours and an additional $3k getting it going. 25hrs was cleaning, repair, and other little maintenance tasks. 75hrs was spent trying to figure out the software madness that controls this sucker. When I say 'Software Madness', I am being kind and gentle. The software was designed by a coder in my opinion - not a software designer. That means that it eventually gets the job done but in an unbelievably strange and non-intuitive way. There is no clear work-flow to follow, buttons and functions are scattered all over. It's nuts. The good news is that it is like a maze - once you find a path through it, it is easy to repeat.

First things first - you need a library of parts just like EDA software. The machine needs to know all the geometric data, values, part numbers, etc. This seems simple enough, but it took many hours to get a sense of how this is actually input and organized in the PPM machine software. Defining parts like QFN is silly. I could go on and on about the crusty software, but in the end I figured it out and it works. I wrote some software and a ULP (script) in Eagle to automatically output the data in a way that it goes directly into the machine nearly ready start placing. It was not easy, but the result is that I can program a PCB in minutes. That should be good. The setup process getting the library going is a big, but ONE time effort. I created about 150 different parts that cover all of my current designs. New parts will be added here and there, but never again 150 in a single shot.

Now that I managed to get the software figured out enough to make a PCB, I will go back to the physical issues like holding small and odd shaped PCBs, custom tube feeders, and custom cut tape feeders. I ordered a dedicated vacuum pump instead of using the compressed air venturi which is loud and wasteful. The monitor and keyboard installation was re-done since the previous owner made a mess out of it. There is a UPS being installed for the whole machine to condition and back-up the machine. Our power is generally good but if power failed during a cycle it could be damaged. Hoping to prevent any power problems from damaging old and hard to find/repair parts. PPM keeps just about everything, but it all costs WAY more than a UPS so what the heck. The PC, as mentioned earlier, has been converted to SSD disk for reliability as well. All the tape feeders have been serviced. Should be ready to roll for real.

So far, I would say that this project is good for what I am doing, but is not for the faint of heart or wallet. All in it will be around $8k plus tons of sleuthing and learning a bizarre system. There is some comfort that PPM has parts and service, the cost of that is really out of the range of this machine. If I did all the refurb work and training with them, it would not be worth it. I would have chosen a different path. I would easily have over $20k into it by now. The DIY effort has saved around $12k, I am ok with that - around $120/hr. It was an experiment from the beginning.

Looking forward, I have a PnP that can place 0201's, BGAs, or whatever with decent speed (rated 3600cph) and great accuracy/precision (.0005"). It is built VERY well, super rigid at around 1000lbs. I can do mixed cut tapes to 56mm with by custom CNC'd JEDEC tray, 50+ tape feeders at once, any size or shape PCB, double sided as well. It can hold enough parts to do 3 different PCB's at once which will save a ton of time in change-over. The goal from the beginning was to do small volumes of a variety of PCB's. My current project has 3 PCB's with a total BOM cost of about $275 or so. To build 100pcs at a time with a local CM would be $27,500 for the parts and maybe $10,000 for assembly (double sided, fine pitch parts, 400 parts, 3 pcbs). Now I can make them as I need them and have total oversight and control not to mention MUCH better cash-flow. I still have to put my own time into it, but I would also have to put a lot of time into getting a CM up and running - kits, organization, inspection, etc, etc.

First part placed.....no fiucials were used so the part is slightly shifted. With the fiducials setup, it really nails it. Even the .4mm QFN's are placed spot on with only the on the fly alignment scanner - Quad Align. Super critical parts like BGA use the upward camera.




For those looking to start a business in electronics, I hope this is helpful. I feel like this machine (as difficult as it has been) will transform my business the same way CNC mills transformed my mechanical design business. Buying a current tech new machine is far more expensive and the hobby machines are not up to the needs of a profitable business.

[rx8pilot]

A new video of the Quad working. Finally got a full program cycle to run. The big delay was designing and machining the special PCB holder for my small and odd shaped PCB's. The Quad is setup for square panels and cannot take a weird or small PCB without a special holder. What I made allows me to use the conveyor system. This allows me to be printed and prepping PCB's while the P&P cycles. The machine can be kept in nearly continuous motion this way.

First full run in slow mode.
https://youtu.be/dUpXYxoDnuA

First full run in full speed mode.
https://youtu.be/n3T007W-IAg

[rx8pilot]

Once again....having a SUPPORTED P&P is awesome.

I am not at the final stages of fully commissioning the P&P machine and of course there are still little bits and pieces that need help. Yesterday, one of the belts for the PCB conveyor (called the 'transport') fell off. They were very loose and the adjustments were maxed out. Also the bearings were all squeaking like crazy - taking then off reveals that they were totally dead.

One call to PPM and new parts are on the way! No researching and hunting. Awesome. I also told them about a problem I was having periodically where the machine thinks a PCB is stuck in the machine. I checked and tested all the related sensors and showed them what I know. There response was to send me a new part (PCB) for comparison to help the troubleshooting process - no charge. If it works, I will pay for it, if not I will send it back to them.

The next awesome thing is that they are sending me a software update which I am VERY VERY happy about. My machine has an early version of the software and it is buggy. There was another 4+ years of development on the software before they went to newer hardware. Knocking out some of the bugs will be really nice.

Goal #1 is to have the machine working all the time.

Goal #2 is to be able to fix things myself when something does go bad.

Goal # 3 is to be able to call someone when it is broken and I cannot figure it out. This goal may be the most important since these machines are so intricate, something is going to break. There is going to be some operational problem. When I can call (for a fee sometimes) and get a solution in FedEx that day....it keeps my business running.

[rx8pilot]

This P&P machine is now commissioned for real work. Finally.

 A million little details. Figuring out most of it by myself because I don't have much documentation and what I do have is so poorly constructed, it serves very little purpose. At this point, it does not matter. I have learned it and fixed it and now it is making PCB's. I get my nights and weekends back!




I designed and machined these pallets for my small, odd shaped, fully routed PCB's. The machine will only take rectangular shapes on the conveyor system. These little things are really great and make it easy to to double sided PCB's.

Blank PCB pallet.


Clamps and supports that attach to the grid. Any shape can be accommodated.


Clamps in place....


PCB's clamped in and ready.

[Kjelt]

Beautifull craftmanship on the clamps and high respect for your perseverance   

[ajb]

Nice!

What's the maximum component height the machine can handle?  Doesn't look like there's much height clearance when crossing over some of the feeder areas, but I suppose the perspective is deceptive. 

[rx8pilot]

It can handle rather tall parts but the limitation is not the clearance between the head and feeders. The limitation is the inlet of the feeders. I have some 1" tall electrolytics on tape and they will not fit into the auto feeder. They will, however fit into the custom cut tape feeder that sits in the matrix tray area.

Learning the operational subtleties is the new challenge. For example, I was placing some sense resistors last night, they are 2512 size and very thin. The machine kept missing the picks with these. When I slowed the movement down and looked closely, I noticed the nozzle was going too deep and fast which was bouncing the next part out of position. I adjusted the pickup height for those to be much more precise and slowed it down a bit. It all works great now. Relatively small issue with an easy fix, but it takes time to figure it out and nail the fix. I have 100 different parts going into the machine and they all seem to need some attention for one reason or another - so the time adds up for sure.

Another unexpected time consumer is part orientation. When I designed my part library in Eagle and in the pick and place machine I did not pay attention to the actual orientation of the parts on the tapes. So what happens is that I have multiple 'Theta' corrections that get confusing. I have one correction that rotates the part relative to the tape to match the rotation called out by the CAD data data that came out of Eagle. While that gets the part on the board correctly, the graphics on the screen are rotated so it looks like it will not be correct, even though it is. Some of the PCB's go into the machine rotated as well, which needs a counter correction on to of the other corrections. My goal is to match up all the library parts with reality so that when I look at the PCB, tape, graphics, etc - it is more intuitive to understand what the rotation should be.

At the moment, I am using double sided tape on the PCB to verify rotation by actually placing a part and looking at it. It works, but I would love to avoid that time consuming process.

[mikeselectricstuff]

Another unexpected time consumer is part orientation. When I designed my part library in Eagle and in the pick and place machine I did not pay attention to the actual orientation of the parts on the tapes. So what happens is that I have multiple 'Theta' corrections that get confusing. I have one correction that rotates the part relative to the tape to match the rotation called out by the CAD data data that came out of Eagle. While that gets the part on the board correctly, the graphics on the screen are rotated so it looks like it will not be correct, even though it is. Some of the PCB's go into the machine rotated as well, which needs a counter correction on to of the other corrections. My goal is to match up all the library parts with reality so that when I look at the PCB, tape, graphics, etc - it is more intuitive to understand what the rotation should be.

Which is fine until you get to parts like SO8's that can come either on tape or in tubes... and of course you don't know which at design time

[rx8pilot]

Smells like experience.  The speed of the machine has to make up for the slow and tedious nature of the setup operation.


This morning, my oven broke. Seems to be a bad fan bearing. Hope that does not take much effort.

Update:
Other than the fans being in the belly of the beast, the fans themselves are easy to get. Digi-Key has the exact model and 866 in-stock for $14. Considering that my oven is a bit dated (Ok industries JEM-310) I am glad to have an easy fix without having to substitute a part. The controller is a Z80 based all through hole design to give you an idea of how far back this goes. It really works great.



Sent from my SAMSUNG-SM-G890A using Tapatalk

[rx8pilot]

Victory today getting 0402 passives down with success! You would think this is an easy and simple step, but it is not. The machine itself certainly has the precision to place these smaller parts, but the feeders are too iffy to get the parts in position to be picked up reliably (there are 0201 capable feeders, I just don't have any). So I decided to use cut tape and put it down on my cut tape holder. Any normal person would expect that if you put a piece of cut tape with 80 components on it, you would simply teach the first and last parts and it would be ready to go.

This is not true at all. The PPM Window software is a maze of oddity and bizarre  that defies explanation sometimes. In general, you can do anything but expect to take the longest path you can imagine - times 10. Over the last 20+ years I have interfaced, designed, maintained, owned, and operated $millions of worth of specialized commercial software/hardware for a wide variety of tasks. In concept, this pick and place machine is among the simplest of the machines I have faced. The learning curve on my 5 axis CNC mills was much faster. Why? Because the software in this case was coded without any design. No consideration for human logic. It's kind of the result you would expect if you designed your PCB before the schematic.

Step 1: Pick up part at specified position
Step 2: Place part at specified position.
Step 3: Repeat until all parts are placed.

Obviously a bit simplified and there are are many questions that machine will need answers to, but that is the basic principle.

So, back to this project. It took about 10-12 hours to figure out how to deal with cut tape. Now that I know how to do it, I can repeat it much quicker, but it is about 1000 times more complicated than it should be. I had to shoot a 20 minute how-to video for myself just so that I would not forget all the bizarre non-intuitive steps needs to make it happen.



So after some serious frustration, it is now capable of doing 0201 and up. I need about 20 more 8mm feeders and may decide to get the 0201 capable variety. In the meantime, I will work through the learning curve and work arounds. PPM generally ignores my emails that have questions and have suggested that I have someone come out for training. They recommend 40 hours of training! 39 of those hours are spent teaching the work-arounds needed.

Some of the responses I get from my questions over the phone: "Why would you want to do that? This is how you do it...." On the surface, it seem innocent, but in the context of the conversation, it means they are not understanding my problem before telling me how it is. The curvy convoluted path needed for every single thing should be a concern for them. How much easier would it be to sell and support a machine that is easy to use and does not require 40 hours of training for every person that will operate it? Like I have said before, I am thankful they are there with spare parts and they have answered some critical questions so far. The disappointment in the software is probably less than what I would have if I was on the original DOS version I guess. At this point, I think they have cut me off unless I spend more money with them.

I have done 99.9% of the repairs and 'training' and I will continue to do that. As my business grows, I will be hoping for a less convoluted option for pick and place.

[LabSpokane]

You must like abuse.  I wouldn't mess with 0201s unless I had a gun to my head and no other options.

[coppice]

You must like abuse.  I wouldn't mess with 0201s unless I had a gun to my head and no other options.

I wouldn't mess with most small SMD devices, but I don't need to. This is why nature created minions. 

[rx8pilot]

You must like abuse.  I wouldn't mess with 0201s unless I had a gun to my head and no other options.

I seem to invite abuse more than I should. Buying a used pick and place machine off of eBay and 'just figuring it out' turned out to be rather self-abusive.

The use of the small high density stuff is driven by the restricted space and weight that I have to work with. I wish I made big rack mount gear with tons of room and big square PCB's. I am in the small box biz.

The good news is that I do all the mechanics so the integration is very tight. At this time, the smallest parts are 0402 and .4mm QFN's and DFN's. Going even tighter will allow a few more features without having to re-invent the enclosure which is a complicated beast.

[coppice]

You must like abuse.  I wouldn't mess with 0201s unless I had a gun to my head and no other options.

I seem to invite abuse more than I should. Buying a used pick and place machine off of eBay and 'just figuring it out' turned out to be rather self-abusive.

Its only self-abuse if its a dead end. If it starts you on a path that leads somewhere valuable its just the normal hard going to get up to speed in a new subject area.

[rx8pilot]

I guess that is true. In the end, I have a pile of new skills and understanding in addition to a rather capable assembly machine. Since I am still well under $10k, it will clearly pay for itself even if I added all my hours in.

The very first run was a great test.....

I needed 5 production boards for a quick delivery. I needed 5 more of a modified new revision to test. I needed 1 with a modified BOM for a special case customer. I did all of them within minutes. That, of course, was after the effort of setting it all up in the first place. It is well on it's way to having about 110 unique parts pre-loaded and ready to place. That allows me to make any of my normal PCB's without a change to the machine. Once I get there, I will declare full victory!

[SteveDC2]

I have owned a Quad 4C/68 for a few years myself and purchased mine for exactly the same reasons. Love working with it.

My machine is the DOS version and the control software that comes with it is very... functional... I did also get a copy of the Quad "AutoProg" program but this was such a complete disaster of a piece of software I gave up after about 2 days of getting absolutely nowhere with it.

Instead, I revered engineered the file format used by the DOS control software and wrote my own conversion tool.

My conversion tool takes in 3 files.

File 1 = a list of various component dimensions and measurement locations (i.e. laser height etc..)
File 2 = a list of pickup locations, including waffle area repeat information for components not in reels.
File 3 = PnP file from CAD software. Currently I support Altium and KiCad directly but others would be simple to add.

The script allows the board thickness and dimensions to be entered and it then calculates all the placement heights so that components are not crushed into the board.

I have not used the machine for about 12 months and have just revived it to find an issue though...

When I make a component measurement by picking up the component then issuing "function 31, measure LAE data" the component offset reported is off by about 80 mils in both X and Y. Anyone have any thoughts on what could cause the component center to be off, and by so much? I expect a few mils but 80 is more than the size of the component and obviously wrong. It then caused the component placement to be off by the same error since the Quad tries to 'correct' for this pickup error.

Great thread though, enjoyed reading everyones experience but wish I had found it earlier so I could have help you through the pain

BR,
Steve

[rx8pilot]

Glad you joined the thread Steve - should do an update soon.

I will not be able to say anything about the DOS version of software since the PPM upgrade removes all of that and has a new interface card to communicate with the motion control and vision. All the parameters and operations are different - even the manual hand controller. On the machine itself - I have repaired, adjusted, inspected, or replaced just about everything. Still learning something every time I run the machine, either something about the machine itself or a detail about the process.

Long term awesomeness would be to find the frames of these machines and convert them to use OpenPNP. The mechanics of the machine are really nice, but the old and crusty software is a burden. At this point, I would say that this machine has been fantastic overall for how I use it. I would be out of business if I was a board assembly house though. It takes too long to get it setup. Like you, I wrote some scripts to program it with very little manual effort. The work for me is the convoluted and bizarre way of teaching the feeders, matrix trays, tube feeders, etc. It's just a strange and confusing approach that is slow. Once it is ready to rumble - all is well. It has been stable and predictable. The only things I have are a few feeders that need help, but that is an easy fix.

Take some photos if you are able. I will try to add some new ones as well.

[H.O]

When I make a component measurement by picking up the component then issuing "function 31, measure LAE data" the component offset reported is off by about 80 mils in both X and Y. Anyone have any thoughts on what could cause the component center to be off, and by so much? I expect a few mils but 80 is more than the size of the component and obviously wrong. It then caused the component placement to be off by the same error since the Quad tries to 'correct' for this pickup error.

Disclaimer: I have zero experience with this type of machine.
What's doing the actual measurment? Is it one of those "laser thingies" or is it your traditional camera/vision setup?

Is it possible that something has come loose and moved so that when the machine positions the center of the nozzle over the center of the camera (if that's what's being used) the nozzle is no longer in the center of the image. Or possibly that the software has lost the calibration/offset number (if it has such a thing)? I don't know how the "laser thingie" works but I'd imagine it too would have to be adjusted and/or have and offset/calibration number applied.

[rx8pilot]

The Quad machines X/Y are linear optical encoders driven by steppers. It is very accurate, reliable and repeatable in general. The Z is a stepper with rotary encoder and Theta is just a stepper I think. Each time it picks up a nozzle it measures the tip and shape of the nozzle with 'Quad Align' which is a side facing line camera illuminated with a laser. Essentially it is looking at the shape of a shadow as the nozzle or part breaks the laser line. It measures the nozzle as it is rotating to pickup runout as well.

The PPM version has an offset procedure to correlate the vision system with the nozzle. I have not yet had anything drift. It is possible that some dust or dirt could be on the linear encoder causing a few skipped steps. Hard to say.

[H.O]

If dirt on the linear scale(s) caused a position error then wouldn't it fail picking the parts up? It's my understanding that it consistently pick the parts correctly but (also consistently) places them 80mils. The fact that it's in both X and Y makes it even weirder. Like I said, zero experience with the machine here, just thinking out loud....

[rx8pilot]

You are probably right. Maybe an offset register like you said earlier.

[H.O]

Another idea... I guess it's using a downlooking camera for fiducials, could that have shifted or had its offset values changed.

[rx8pilot]

Either that or the up camera is being used and not offset correctly. If either is the case, should be an easy fix. Both of my cameras were off when I first got it. After getting the offsets dialed in, it has been tight since. The Quad Align camera also needs some inspection - it could be sending incorrect offset data after it scans the part.

[WhiteMatterLLC]

I have thoroughly enjoyed reading about your experiences with the Quad 4000C! Thank you so much for documenting this for future users.

Our company is purchasing a Quad 4000C (with Windows XP) and I am in charge of figuring out  how to get it to place 0201 components.

I am trying to track down a user manual (PPM wants over $100 for it...) and training videos. If anyone has these and would like to share, I would be very appreciative!

Sent from my Nexus 5 using Tapatalk

[rx8pilot]

I have the programming manual that they emailed to me for free. I do not, however, have any documentation of the hardware or the machine setup, calibration, etc. All of that I had to figure out on my own with substantial pain and suffering. The software can do what needs to be done and it is stable - but it is about as intuitive as a 100,000 piece jig saw puzzle. I shot 'reminder' videos of most of the programming that I have to watch regularly to remind myself how to setup the machine. The programming manual is so poorly organized and is missing many of the diagrams. It is definitely helpful, but you need to be a hacker to figure it out. 

YouTube Videos that are helpful. Painful, slow, scattered, but helpful.





For 0201, you will need at least the slim feeders and the precision 0201 nozzle. The precision 2mm pitch feeders are really the way to go from an outfit in San Diego.
http://www.ebay.com/itm/121716883289?_trksid=p2060353.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT

PPM sells a 2mm pitch precision feeder but it is more expensive than the value it provides. As much as possible, I want to support PPM because they are needed to keep these machines running. The problem is that they still have the parts and service rates that were appropriate in the 90's. My interest in this older machine was the low cost of entry into a pro machine. I have less than $10k in it total. If I went to PPM for every detail and training, it simply would not exist at all since the cost would have tripled. Once I got to the bigger numbers, I would have been keen to look at newer machines.

Where is your machine coming from? Where are you located?
Feeders? Options? Are you doing in-house assembly for engineering support or for an assembly business?

[WhiteMatterLLC]

Thanks for linking your reminder videos! You mentioned them in a previous post, so it's nice that you are sharing those as well.

We are picking up the machine in North Carolina and we are located in Seattle, Washington. It is for in-house assembly work, since we are at a point that human percision is just too tedious and prone to mistake. We would normally just pay an assembly shop to do the boards for us, but we are in a situation that that becomes too costly. Small volume, high precision, can't use stencils to place solder, plus we would like the ability to swap out some components to change the amplifiers for certain customers.

The company selling us the machine has it up and running as we speak, plus they have offered to give me a crash course on programing it. Hopefully this means I can have this baby up and running quickly! As for feeders, they are offering to give us as many feeders as we need for our small operation, given that they don't need the ones we want. From reading your blog I have gathered that 0201 components work better with a the smaller indexing feeder, but they still work with the normal 8mm feeder. Is that correct? And as for nozzles, we plan on purchasing them from Count on Tools. http://www.nozzles4smt.com/Quad-0201-Nozzle-with-906-Ceramic-Tip_p_1598.html] [url]http://www.nozzles4smt.com/Quad-0201-Nozzle-with-906-Ceramic-Tip_p_1598.html [/url]

Being a start up company, we too are looking to save money wherever possible. One of our reasons for choosing this machine was that it is still supported by PPM, and they have been very poractive in communication with us. We would like to support PPM as much as possible, but I guess we will just have to see how expensive they are relative to the free market.

[rx8pilot]

The videos are made from PPM - I have not published my videos yet.

The challenge with small parts on tapes is not the indexing distance - but the repeat-ability. The standard 8mm feeders only get the part close to where it should be, not exactly where it should be. The 'slim' feeders are still 4mm indexing but much better repeat ability and I use them for 0402 with no issues. For 0201, they will probably pick most parts, but there will be more chance of mis-picks. For reliable, 'works everytime' 0201 you will need the SlimPRO or ThinPRO feeders that have tighter mechanics. The system has no way of knowing where the parts actually are, you tell it where they should be and that that is were it will pick up every time. If you don't pickup close to the center of the part, they tend to go crooked and the Quad Align will fail. When you think of how small the 0201 is, the target to pick is very small. I am saving my money to load the machine with as many 0201 feeders as I can afford. Right now, my design choices are dictated by the limitation of the the P&P I have.

Take a look at the JEDEC variable width cut tape holders I made. They allow me to do low-volume 0201 perfectly, but keeping the tapes loaded is not easy. Real feeders are the way to go.

It is for in-house assembly work, since we are at a point that human percision is just too tedious and prone to mistake. We would normally just pay an assembly shop to do the boards for us, but we are in a situation that that becomes too costly. Small volume, high precision, can't use stencils to place solder, plus we would like the ability to swap out some components to change the amplifiers for certain customers.

That is pretty much my story. Having  batch of 20 PCB's done was VERY expensive and time consuming. My boards are almost all double sided, fine pitch, and low volume. The organization, kitting, setups fees, lead-time, and lack of flexibility is a deal killer. The P&P process (machine, oven, printer, tools, skills) overall are a total PITA - but less than outsourcing a variety of prototypes, pilot runs, specials, etc. After the blood, sweat, and tears - it was a great decision.

Random notes:
Do not underestimate the challenge of printing paste. It is a very unforgiving and critical part  of the process. Re-work of fine pitch parts will quickly eat up the time saved by automatic placing. You have to nail the printing which starts with a good stencil design and ends with delicate skill. Anyone can do SOIC and 0805 blindfolded - fine pitch rapidly gets more difficult and the penalties get steeper.

McMaster Carr is your friend. PPM has bearings and every other part - but McMaster is a fraction of the cost for literally the exact same part - brand/part number.

The power plugs on the feeder banks can be plugged in backward - don't do that. It's bad.

The rollers and belts should be fresh to work well. They are very easy to replace and not too expensive - I got them from PPM. The metal rollers get gunked up and sticky. They don't need to be replaced, but removed, cleaned, greased. If they are sticky, the tape will not peel and you will regularly scream the darkest of profanities.

Clean the contact strip on the feeder bases, they get dirty and the the feeders 5v control section does not like anything to get in it's way. Not sure why the 5V power rail is routed all over creation and it's lossy once you get a full rack of feeders in place. If the voltage gets below 4.75V at the feeders - the reliability goes down.

I cloned the HDD and replaced it with an SSD with IDE interface. I also clone that regularly with Acronis to a USB and the cloud. re-configuring the system with XP and all the proprietary drivers is not an experience I am willing to have. Learn how to backup the data base of programs and parts off the computer as well. With this system, you don't save files - it all goes into a database that you will learn to hate but have to live with.

Keep sharp objects, poisons, and firearms at a considerable distance when you are going through the learning curve. You will say WTF! about a million times since nothing at all in the software is intuitive. 

Post picks and experiences - I will help if possible. Congratulations - it will be a great tool to have (after you learn it)

[mikeselectricstuff]

Hopefully this means I can have this baby up and running quickly!

Nobody, ever in all history has got a pick & place machine & full assembly process up and running quickly, unless they've done it before.
It's not just about the machine and all its quirks, but all the little details of all the processes, any one of which can bite you.
As mentioned before, there will be frustration and probably swearing before everything is going smoothly.
 

[WhiteMatterLLC]

Thank you for all of the notes! It's probalby a good thing that I don't have access to any high balconies. It's probably a little masochistic, but I am really looking forward to the challenge of getting this PnP machine working.

I think we have budgeted a decent chunk of money for feeders. It's too bad that they are so expensive, but if it means that production is easier and more repeatable then it will be worth the expenditure.

Did you end up using a vacuum system over an air compressor? Noise is a concern, plus we would like to take up as little space as possible.

[rx8pilot]

It's probably a little masochistic, but I am really looking forward to the challenge of getting this PnP machine working.

With all the pain and suffering - I did enjoy the process. I really walked in blind - not knowing the process or the machine. To make it more fun, I purchased a broken machine that I had never seen and had no practical access to manuals. That is a LOT masochistic for sure.

Did you end up using a vacuum system over an air compressor? Noise is a concern, plus we would like to take up as little space as possible.

I built a vac system. 80/20 framing, Gast pump, electrical on DIN rails, valves, tank, gauge, etc. Its quiet and does not need to run much since it has a tank to hold vac. The air compressor is a 1.5hp California rocker type that is about as quiet as a piston compressor can be. The machine only needs a tiny volume of compressed air for the puff off. The size was chosen to be capable of making venturi vacuum if the primary vac system fails - I love backup plans. I also use it to clear the apertures on the stencils.

On top of all my other backups, I have the entire P&P on a big rack mount UPS for filtering, leveling, and power outage protection. The electronics are old and expensive, clean and stable power gives me a better chance of survival.

[rx8pilot]

There are a number of pics that I never put in this thread.


The PC rack slide was never installed correctly and was destroyed. I re-mounted it and cleaned up all the cables in the process. Re mounted the monitor/keyboard too.


This is another pic of the PC mounting repair and you can see the vacuum rig on the floor. I put considerable effort into the vacuum so that I would never have to worry about it. 80/20 framing, and all high-end industrial parts.


Not quite finished vac pump system.


CNC machining the PCB holder clamps. Without the PCB pallets, I would not be able to do small and odd shape PCB's. Deal killer.


Cut tape JEDEC tray holders coming together. I would make some improvements if I did these again, but overall not bad. Saved me from needing to by parts on reels and a lot more feeders.


Test fitting the tapes.


Any width tape in either 6in or 12in lengths. A real problem solver for short runs.


Loaded in the machine, ready to roll.


The software was not really ready for these holders so I had to get creative. The matrix trays are defined in a config file that is somewhat hidden. I found it and learned the syntax of the config well enough to add my own custom trays. This is a critical stop-gap to be able to do prototypes and 0201 parts that I don't have feeders for. They were not easy to program and align, but I am very thankful to have them.

[l0wside]

Oh and as far as splicing...  I would never pay $600 for a tool that didn't do the job almost completely automated. 

Well, to be fair, the tool is half that. The rest of the cost is the consumables ranging from 8mm to 56mm. Seems simple enough to deal with flat paper tapes but the deep pocketed plastic tapes for bigger components like inductors and push buttons seem harder. A website called http://www.nozzles4smt.com/ has a cut tape carrier that is adjustable to use various width tapes. Looks promising for my kind of work. They also have other consumables for Quads - nozzles, rollers, etc.

How do you handle this issue now? Have you purchased the splicing tool (600$ is not exactly cheap), or have you found a different solution?
I have one component (QFN36 in plastic tape) which is not available from the usual suppliers, but way too expensive to purchase a full reel.

Max

[rx8pilot]

How do you handle this issue now? Have you purchased the splicing tool (600$ is not exactly cheap), or have you found a different solution?
I have one component (QFN36 in plastic tape) which is not available from the usual suppliers, but way too expensive to purchase a full reel.

Max

It did not take long to figure out that I can load a tape with no leader at all. I just have a kapton tape dispenser with 3 widths to connect the cover tape to the end of the previous tape - quite easy and I never lose the first part. The end of the tape is a different matter since the feeders cannot advance all the way to the last pocket. For resistors and capacitors, it's not a big deal to lose a few. With $6-8 parts from Linear Tech - I am more inclined to solve the problem.

What I do on those as a work around is remove the last few parts on the cut tape and line them up next to my feeder loading station. After a some parts have been placed - I take out the feeder and manually back it up putting in the parts that were previously at the end. It is a little fiddly, but allows me to push forward without the expensive splicing tool. The first few times it felt like a major problem. Now that I have done it a whole bunch - I am quick. At this point, I have better places to put $600. It would take a long time for that tool to make up the time needed to justify it.

The next big effort for me is to get the vibe feeder setup to do a variety of tubes. I am still manually placing all parts that come in tubes which is a pain, especially since I have the vibe feeder already.

[ElkoShop]

Last week I picked up two Quad 4c assemblers. (DOS Controller)  I have them both powered up, both home on power up, but I cannot get the HHT to move the head on either machine.
The only thing I can seem to do is home the machine with the HHT.
Is there a sequence to enable head movement?

Thanks in advance.
Steve

[rx8pilot]

The Windows controller is a total do-over - no relation to the original DOS setup as far as software is concerned.

You put two of these things in your house? Starting a business?

[ElkoShop]

I was hoping to get a reply about the HHT from one of the dos guys. Posted here because I hoped one of them might be still be following this thread.

Nope, not starting a business, one machine is heading to Fl (after I can get the kinks out) and the other one (along with extras) will probably be sold.
Have a friend down there and we have been building some small boards by hand. These machines used to share a controller, so the extra won't have one.

I took a non-functional one apart 6 years ago and made a cnc router out of it.

Steve

[Kjelt]

I did not want to continue to hijack the Neoden topic discussing the Quad, so continuing here 

So what is it, first you state that if the Neoden is fully supported by OpenPNP you would switch , which I doubt since it is not as robust as a professional machine you now have, then you say that only a professional $100k+ machine will be able to replace it. If you have nly look at the feeders the Neoden is in a total different league IMO.

I should clarify....
I like the commercial elements of my current setup with the Quad. Sturdy and clearly designed for 24/7 use. Nothing to complain about...until I get to the maintenance or replacement part. In general, I cannot afford a $100k replacement but I also hesitant to pay big money to keep an old machine running. I am on the lookout for a fairly low cost replacement and it does not need to be amazing - only better than what I have in terms of maintenance costs.

My usage is infrequent, but when I need PCB's I need them NOW. So, like I said, as time marches on, the challenge of keeping a Quad running will increase. At some point a low-cost replacement will be on my radar. If my little biz grows to needing PCB assembly all the time - a high-end machine may make sense. As I am configured now, I can only consider the low cost options and accept the various limitations that come with that.

I only got into the Quad because it was a low-cost of entry and I have been able to expand over time to increase capability. There was never an option to go with a modern, high-end system.

I understand. Since you are someone with many talents mechanically as well electronically, did you think about an overview to look what it would take to adapt your (or extra other)  Quad to OpenPNP and 2017 hardware (USB camera's , closed loop steppers or servo's, Win10) etc? Would that not be a viable option towards an entirely new less quality P&P machine ?

[rx8pilot]

Hard to say. It seems that it would be fairly straightforward with the exception of the Quad Align system. I have no idea how that sub-system works or communicates with the control system. 6

Considering that I now have a huge assortment of feeders - 230 tape feeders, 6 custom strip feeders that can accommodate 22pcs of 8mm tape each, Vibe feeders that I have customized to take any tube, plus JEDEC trays. It has grown into a rather serious setup but that does not quite help me forget that it is all quite old electronics. If the ancient vision system has an issue, it may be a very expensive thing to fix or replace.

It is very nice to able to load up 80 parts at a time while still having pre-loaded feeders to swap jobs fast. The small Neoden N4 class machines simply cannot handle that many parts (on top of other limitations). Also, it seems the entry level machines struggle with 0402 and smaller. My last design used 30 different 0402 parts and I am hoping to move to 0201 as well.

At the moment, I am trying to plan for an unexpected failure. If I was going to convert the machine to OpenPNP with updated motion control - I would buy a non-working Quad and start from there. Probably swap the steppers with servos (which I already have a bunch of Maxon motors and drives that would probably work) The machine already has glass linear encoders for X/Y.  Would just hope to work out a solution for the side scanning laser which is one of the best features of this system.

I have no intention of replacing this system until I am pushed to do so. At the moment, it is getting the job done.

[PrintTec]

Dear RX8pilot,

We can do 0402 with Neoden and even 0201!

Best regards,

Robert Joosten
http://www.printtec.nl/contents/en-uk/d596_NeoDen_pick___place_machines_with_vision.html

[rx8pilot]

That is good news. I feel like I would need at least 2 (maybe 3) of the N4's to get my jobs done in the same time because of the limited number of feeders. By then, I am up to >$30k plus feeders at which point I would be thinking of a used commercial machine that is still supported.

For 0402/0201 - is it safe to assume you have to image each one of them with the up-facing camera? What is the typical error rate for mis-picking 0402/0201? Is it able to pickup heavy inductors? Tall parts like electrolytic caps?

[mikeselectricstuff]

That is good news. I feel like I would need at least 2 (maybe 3) of the N4's to get my jobs done in the same time because of the limited number of feeders. By then, I am up to >$30k plus feeders at which point I would be thinking of a used commercial machine that is still supported.

For 0402/0201 - is it safe to assume you have to image each one of them with the up-facing camera? What is the typical error rate for mis-picking 0402/0201? Is it able to pickup heavy inductors? Tall parts like electrolytic caps?

Any P&P would have to image 0402/0201 parts - no way these could be done reliably without
N4 can't do tall parts die to the see-saw head design.

[rx8pilot]

Any P&P would have to image 0402/0201 parts - no way these could be done reliably without
N4 can't do tall parts die to the see-saw head design.

This is a distinct advantage of the Quad system - it uses a side scan that is integrated into the head. That allows it to image and apply correction while it is moving to the placement location. Super quick and theoretically accurate enough for 01005 assuming you have the right nozzles.

The downside is that the parts are obsolete for the most part so maintenance can be a challenge.

[solarbotics]

I picked up a Quad 4000C from them a few years ago, and they pointed me to a few of their Youtube training videos.

I downloaded and watched them and created a time index for them. If anybody is interested, I can see about compiling them up into a useful link/post.

Regards,
Dave

[rx8pilot]

I picked up a Quad 4000C from them a few years ago, and they pointed me to a few of their Youtube training videos.

I downloaded and watched them and created a time index for them. If anybody is interested, I can see about compiling them up into a useful link/post.

Regards,
Dave

This is good news - the raw video is marginal if I am being generous.

I shot a whole bunch of video for setups, but never posted it to my Factory400 channel. There are so few people that have these machines configured in the PPM 4000C. Do you still have your machine? Do you do much small part placement 0402 and smaller?

[solarbotics]

I'm replying from that very machine now (glad the terminal can run RDP to my proper workstation).

I don't place 0402 with it - mostly 0804 and some 0603. I think I could do finer, but I'd have to fix my out-of-spec runout on my nozzle first.

https://drive.google.com/open?id=1gj5nM0AzDsevxXsaP46LfILeit25iCpN

Hope it helps somebody else! I've made good use of the video index a few times now.

[rx8pilot]

ooooh...those videos are so very painful but ultimately helped me get setup and running.

[solarbotics]

I just finished decyphering the matrix tray setup video too. Not that bad:

0:00:00 - Introduction to 4C Custom tray Mount Setup, cleanup tabletop
0:02:51 - Start introducing User device

    Program Definition / "User Devices" button under "Customer column"
    00:04:01 - "Machine Configuration" to find new tabletop items?

00:06:15 - "Matrix Tray holders Tab "CustomMatrxTrayMount4C" ..  2Pos" - case sensitive

    Lets you configure you the MOUNT / HOLDER (not the tray)

07:45 - Bottom - "Tray Mounts" configures tray locations
08:15 - Matrix tray Holder options defined
09:50 - Fiducial locations on the Tray HOLDER

    Used to easily use camera to find corners

11:23 - Tray mount position (UL/UR/etc)

    X/Y from the Tray Holder 0,0

19:22 - Configure up custom tray

    "Matrix Trays" tab under "User Defined Devices"
    Configured from X/Y from defined reference corner (top left, right/etc)


    Teach Tabletop Devices - This is where we are going to put whatever matrix tray HOLDER into the machine. It establishes it's XY coordinates.
    Trayholders Tab - Add in the tray you want to put onto the holder
    Select any pocket (highlights) and move to it to see if it's correct
    Somehow Ref1 and Ref2 (Table objects) refer to the fiducials drilled into the matrix tray mount
    HAHA! Right click on a pocket (TL, BR, BL) and assign them as references REF #1, 2, and 3. Pick them off with the controller, and it's aligned!

When filling trays (teach tabletop devices / User / Matrix trays) right click and use popup menu to establish what's in there (picked, empty, disabled, whatever). FILL - totally fills the whole thing.

[rx8pilot]

When filling trays (teach tabletop devices / User / Matrix trays) right click and use popup menu to establish what's in there (picked, empty, disabled, whatever). FILL - totally fills the whole thing.

That is a super annoying thing.

My big challenge was to create matrix tray 'strips' to accommodate my custom holders that can have any size tape in them. It took a while to figure it out, but in the end I can have any mix of tapes on the table.

[solarbotics]

I'm having a hell of a time resolving how to omit one image in a panel that didn't pass QC.

I'll have a "reject" fiducial marker image near the panel, and the machine will detect it, but then will totally reject the WHOLE board, not just that one image (1 of 10 on the panel).

Sooo, how else to disable populating a bad image? Sure, I can go to "Panel Configuration", sort by image, group-select everything in that image and "disable" them, but then toggling them back on means I have to catch all the optional "off" elements in the image I usually have off.

In "run program", I can't do the same "sort by image" (and simply toggle it to "placed"). Closest I've managed to come to is "sort by placement", then scroll through the various nozzle changes and find those image placements. TEDIOUS.

Anybody else have a method to temporarily turn off population of one image in a set?

[rx8pilot]

Anybody else have a method to temporarily turn off population of one image in a set?

I run them from the TEACH PLACEMENTS window where I just select the ones I want. It is not too terrible, but enough that I tend to avoid panels with X-out sections.

There is little question that the PPM software is marginal if I am being nice about it - my workflow is all about working around the bad decisions or lack of features in the PPM software.

As you pointed out.....using partial matrix trays is stupid. The system makes it so difficult/tedious to skip empty pockets. The RUN mode offers very little control to deal with empty feeders or mis-picks - so I run a ton of boards from the TEACH mode where I can control the process a lot more.

Either way - I still believe this system is leaps and bounds above any of the low-cost Chinese offerings and a fraction of the cost of a new-ish commercial option. I have slowly learned how to get boards assembled with it and now rather quick setting up and operating the machine.

[solarbotics]

Yeah, that's sorta what I do to. Kinda an oversight not to let us sort by clicking column headers like any modern software. I should call them and see if their software has been updated at all lately.

A few years ago I sent them a little heart-felt diatribe telling them that as a SMALL company, they should open-source (at least) their feeder schematics and gain a toe-hold in a new and up&coming field. Told them how 3D printing made CAM much easier, and they should consider opening their standards and let the OSH world adopt them and use their feeders as a template. Never got feedback on that. 

Further to that, YES I wish their software was nicer to play with. I have NO idea what database format this is, and wish that standard was published.

[rx8pilot]

Yeah, that's sorta what I do to. Kinda an oversight not to let us sort by clicking column headers like any modern software. I should call them and see if their software has been updated at all lately.

Nope. Even if it was, they would be asking some insane amount of money for it.

A few years ago I sent them a little heart-felt diatribe telling them that as a SMALL company, they should open-source (at least) their feeder schematics and gain a toe-hold in a new and up&coming field. Told them how 3D printing made CAM much easier, and they should consider opening their standards and let the OSH world adopt them and use their feeders as a template. Never got feedback on that. 

They have all been very nice and helpful to me, but the business appears to be rooted in times gone by. The pricing for used spare parts is enough that if my machine had a couple of significant failures - I would just dump it and look at other options. There is no way I would put many thousands more into this machine to replace the vision system or some other high-value sub-system.  I am going to assume that they would rather see these machines slowly fade with them than do anything to push them along. I think you are more likely to win the lottery three times in a row than see PPM go that direction.

I have even considered making my own feeder control PCB - it would be cheaper, easier, and higher performance than getting the schematics for a design that goes back to the early 90's.

[JeremyCanary]

Hey All

Thought I would jump into this thread. I'm an EE at a small electronics company with a whopping 17 Quad machines of various models. Aside from EE related duties, I also keep these workhorses running (or atleast attempt to) for the operators on our production line.

I'm relatively new to working on these machines, but i'm sure we can all teach each other a thing about running these interesting and often PIA, pick and place machines.

-Jeremy

[rx8pilot]

Hey All

Thought I would jump into this thread. I'm an EE at a small electronics company with a whopping 17 Quad machines of various models. Aside from EE related duties, I also keep these workhorses running (or atleast attempt to) for the operators on our production line.

I'm relatively new to working on these machines, but i'm sure we can all teach each other a thing about running these interesting and often PIA, pick and place machines.

-Jeremy

Awesome! Glad you chimed in.
Are any of your machines using PPM software on Windows 7/8/10?

My biggest area of fussiness is the feeders with small parts. I have some of the precision feeders and in the process of modifying the standard 8mm to be precision feeders.
I need to place mainly 0402 passives now - going from 0805....to 0603.....and now 0402 as my designs have become more and more dense. If all of my 8mm feeders were perfect enough to do 0402 all day long, I would be super happy.

[JeremyCanary]

No, none of our machines are on windows 7 or later. They're all running XP. I did find some documentation about windows 7 stashed away when I assumed the responsibilities of the person before me. Perhaps the person before me had plans to experiment with later iterations of windows, I think i'll look into it.

As for parts in the 0402 package type, I believe one of our products has a few 0402 resistors and/or caps (I can't remember off the top of my head). I can double check when I pop into the office tomorrow morning. I do know that the our parts database in the qsoft software has an 0402 profile. We have a few precision feeders as well.

[rx8pilot]

My system is XP as well - just forgot to put that in my question.

As far as I have learned -there is no benefit to later versions of Windows. Qsoft is still Qsoft regardless of the OS version.
Do any of your machines use the old MS-DOS software/hardware?

[JeremyCanary]

At one point they were. We invested a pretty penny getting them all upgraded from DOS to XP.

[JeremyCanary]

I'm curious to see how your modification to the 8mm feeders works out in order to place 0402 parts! Keep us updated!

[rx8pilot]

I'm curious to see how your modification to the 8mm feeders works out in order to place 0402 parts! Keep us updated!

For sure!
I have no plans to make a business out of them, but if I can offset the R&D by selling off a few - I would do it. My goal is to do about 60 or so once I get it dialed in.

I have to CNC machine the tape sprocket, motor encoder, modify the main casting, and add a gear tensioner. In batches - it is not too bad. Having a reliable capability to do 0402/0201 parts will be amazing. I already have the designs, donor feeders, tools, and a Haas Super Mini Mill that barely gets used.

[jmelson]

I've been looking to upgrade my Philips CSM84 P&P, it is not accurate enough.

I need to do some TSSOP and QFP parts with 0.5 to 0.65 mm lead pitch.  Also, the centering jaws on the CSM seem to only handle parts with the right width to length ratio, square parts are not accurately placed.

I've been offered a Quad 4C that might be in my price range.  But, I'd like to hear more about how well your machine is working for you now that you've had time to run it.  How is the vision programming, how accurate is it on the fine stuff, how reliable is it as to the machine, and also the picking reliability?

Thanks much,

Jon

[rx8pilot]

Mine runs on Windows XP through Windows 10 - although I have elected to leave it on XP and currently disconnected from the network.

It is a rather precise and reliable platform. The Quad Align system is really nice as it is able to scan the part while it moves from the feeder to the placement position. It really has no issues with fine pitch components at all. The up looking camera rarely gets used, but when you need it you really need it. I have placed very fine pitch parts without any issues. It is stepper based motion with fast ball screws, but uses glass scales for very repeatable performance.

The feeders are a mixed bag. The quality varies considerably over the years. I have about 250 feeders or so and find that they regularly require some TLC. For smaller parts like 0402 and 0201 you have to have the slim precision feeders, but they are also not all created equal. I have about 25 precision feeders that I have worked on enough to get solid reliability. The others get used for larger parts that do not need the precision.

I have a modification that I am doing to the standard 8mm feeders to make them very precise and planning to make about 15-20 of those over time as I need them. They are capable of 01005 parts if i spring for the very expensive nozzle needed for those parts.

Overall, I have been very happy with this system for the money and time it has required. It is such an unforgiving process and nice to have a solid placement machine. Like any other P&P, it requires a lot of attention up front and a person that can recognize the myriad of subtle details that make the difference between success and failure. Since you already have the experience, you undoubtedly know what I am saying.

The financial low-down: Mine was initially $5k with about 50 feeders if I recall correctly, but broken and not workable. I slowly repaired the system and taught myself how to use it. At the time that I placed my first part I had put another $1000 into it. Over time, I purchased more feeders and machined various accessories and replacement parts. As it sits today, I probably have about $7k or a little more in it. It is FAR more capable than a Neoden or whatever other Chinese build-to-the-lowest-price special that cost a lot more than that.

The PPM software eventually gets the job done but is really bizarre. I have heard the DOS version is actually better in some ways although I cannot say myself.

Hope this helps.

[jmelson]

I've had a few phone conversations with PPM.  They said something about the 8mm feeders taking up two slots.  Does this mean a machine with 98 feeder locations can only take 49 8mm feeders?  Yikes!  The Philips CSM series, Samsung CP and SM series and a few others have standard feeders that mount on 16 mm centers.  If this is true, then it sure is going to send me away from any machines using Quad feeders.  I hope you can clear this up.

Thanks,

Jon

[rx8pilot]

I've had a few phone conversations with PPM.  They said something about the 8mm feeders taking up two slots.  Does this mean a machine with 98 feeder locations can only take 49 8mm feeders?  Yikes!  The Philips CSM series, Samsung CP and SM series and a few others have standard feeders that mount on 16 mm centers.  If this is true, then it sure is going to send me away from any machines using Quad feeders.  I hope you can clear this up.

Thanks,

Jon

There are 2 different types of feeder bases - standard and slim. The 8mm feeder capacity is based on the standard bases.

Toward the end of the production run on these machines there were two narrow 8mm feeders introduced that would fit in the standard width slots, but they also made special feeder bases that took advantage of the the slim feeders, allowing them to be closer together and fit more feeders on a given base.

The standard 8mm feeders will fit on these slim bases, but take up two slots.

Standard 8mm feeder on standard feeder base: No space in between. Wider feeders.
Slim 8mm feeder on Slim feeder base: No space in between. Narrow feeders adding about 25-30% more feeders per side of the machine.

Standard 8mm feeder on Slim feeder base: Wasted space between feeders, fewer total feeders than on matching standard feeder base.
Slim 8mm feeder on Standard feeder base: Wasted space between feeders, same number of total feeders regardless of standard/slim being used.

[jmelson]

There are 2 different types of feeder bases - standard and slim. The 8mm feeder capacity is based on the standard bases.

Oh my!  It is a lot more complicated than I thought!  Yikes.

Well, thanks very much for explaining it.

Jon

[rx8pilot]

I am almost due for the "5 year experience" video on this thing.

So much learned. So many tears. So many circuit boards.

[jmelson]

Does the Quad IV C not have wheels?  I notice in the pictures for the auction that all the Quads are sitting on wooden blocks.
(Maybe they didn't want to put flat spots on the wheels.)

Thanks,

Jon

[rx8pilot]

Does the Quad IV C not have wheels?  I notice in the pictures for the auction that all the Quads are sitting on wooden blocks.
(Maybe they didn't want to put flat spots on the wheels.)

Thanks,

Jon

The bottom of the machine has 1/2-13 threads so you can put whatever you want on the bottom. For economy, I put adjustable leveling pads on mine so that it is very stable. Operating directly on wheels is probably not a great idea. The gantry acceleration and deceleration is quite high and the machine would likely wobble/shake/move.

I took the time to precision level mine and checked for any twist in the gantry rails - the same way I deal with my large CNC machines.

[jmelson]

The bottom of the machine has 1/2-13 threads so you can put whatever you want on the bottom. For economy, I put adjustable leveling pads on mine so that it is very stable. Operating directly on wheels is probably not a great idea. The gantry acceleration and deceleration is quite high and the machine would likely wobble/shake/move.

I took the time to precision level mine and checked for any twist in the gantry rails - the same way I deal with my large CNC machines.

Interesting.  My Philips CSM84 has both decent wheels and leveling feet.  I thought most other machines also had both, wheels to roll it in and feet to stabilize it after putting it in place.

Well, I had the opportunity to get a Quad IV C or a Quad QSA-30, and I went for the QSA-30A because it was newer, and looked pretty complete and clean.  The IV C machines all looked a bit dirty, and I wasn't sure they were complete.  The QSA-30 was made by Samsung, but is fitted for Quad electrical feeders instead of Samsung's older mechanical feeders with air cylinders to advance them.

Well, it is going to take a while to get all the accessories to put this machine into operation, but I got it for a SONG, so I have money to spend on feeders and such items.

Thanks,

Jon

[rx8pilot]

Does it have the PPM Windows software or the original DOS application?

[jmelson]

Does it have the PPM Windows software or the original DOS application?

Well, I don't know!  I am GUESSING it is the original, but have no way to know.  Does the DOS or Windows software run on an embedded computer built into the machine, or is it a separate computer?  Almost anything that is not permanently attached to the machine has been removed.  It does have the hand-held controller and 3 nozzle changers.  No nozzles, no video monitors, keyboard or mouse.  and, I suspect no manuals or anything else.

Thanks,

Jon

[rx8pilot]

Does it have the PPM Windows software or the original DOS application?

Well, I don't know!  I am GUESSING it is the original, but have no way to know.  Does the DOS or Windows software run on an embedded computer built into the machine, or is it a separate computer?  Almost anything that is not permanently attached to the machine has been removed.  It does have the hand-held controller and 3 nozzle changers.  No nozzles, no video monitors, keyboard or mouse.  and, I suspect no manuals or anything else.

Thanks,

Jon

The software runs on a standard PC in a rack mount enclosure. If it is missing, I am not sure how you would get the hardware and the custom interface ISA cards along with the software and the calibration data. Could be quite a challenge unless you get PPM involved, at which time the machine will get very expensive even if it was being delivered for free.

Take a look at the beginning of this thread - you will see where the PC typically is installed.

[jmelson]

The software runs on a standard PC in a rack mount enclosure. If it is missing, I am not sure how you would get the hardware and the custom interface ISA cards along with the software and the calibration data. Could be quite a challenge unless you get PPM involved, at which time the machine will get very expensive even if it was being delivered for free.

Take a look at the beginning of this thread - you will see where the PC typically is installed.

Wait, the subject line is now WRONG!  I ended up buying a Quad QSA-30A, not the IV C machines.  Samsung CP machines have a "headless" VME motion/vision controller, which I think is based on a 68K processor, and a standard PC motherboard mounted to a plate inside the box.

The QSA-30 is made by Samsung for Quad, and sure LOOKS like the rest of their CP line.  The big difference is it uses Quad electronic feeders instead of Samsung mechanical ones.   But, I've never gotten a good look at the interior with the covers off.  So, I don't knmow if the computer setup is the same or different.  There are some indications that there is just ONE computer, running Win 95, that does both the MMI part and the motion/vision part.  I'm trying to nail down all these things that I don't know about this specific model.

Also, it looks like I might have to fly down to Austin, TX and rent a truck to haul it back to St. Louis.

Thanks,

Jon